River Dee Hatchery Appraisal
Final Report
October 1st, 2016
This report has been produced by the River Forth Fisheries Trust with Trex
Ecology, the University of Stirling, The Queens University of Belfast and Aqua
EcoSystems BV.
River Forth Fisheries Trust: River Dee Hatchery Assessment
ii
Report Authors:
Tommy McDermott – Trex Ecology and River Forth Fisheries Trust*
Alison Baker – Forth District Salmon Fishery Board
John Bostock – University of Stirling
Colin Bull – University of Stirling
Chris Clayburn – Aqua Eco-Systems BV
Joanna Girvan – River Forth Fisheries Trust
Paulo Prodöhl – Queens University Belfast
*Primary author
River Forth Fisheries Trust: River Dee Hatchery Assessment
iii
Executive Summary The River Dee is currently undergoing a decline in the number of rod caught Atlantic salmon. These
declines have been sufficient to lower the long term average in catches, a key indicator of a downward
trend. As part of an investigation into arresting this decline, the Dee Board and Trust commissioned
an independent study into the feasibility of stocking Atlantic salmon to boost rod catches to three
levels, 100, 1000 and 5000 additional rod caught fish.
This study developed a three stranded approach to assess the issue. Firstly, it reviewed the biological
information from both the published literature and from the distribution, abundance and habitat
uptake of juvenile salmon on the Dee using available habitat and electrofishing data. This strand also
aimed to identify to most suitable life history stage to stock. Secondly it costed a ‘paper’ hatchery
build to produce sufficient numbers of juveniles to deliver the required increase in the fishery, and
carried out a cost benefit analysis of that operation. Finally, it reviewed the legal frameworks, such as
the Special Area of Conservation Designation on the Dee and the obligations of the Dee District
Salmon Fishery Board and the National Park, and identified other management options which could
deliver the required increase in rod caught fish.
Understanding the current population abundance is key when discussing stocking and it is difficult to
robustly assess the numbers of returning fish to the Dee. The literature shows that catch has been a
poor indication of total population sizes in Atlantic salmon. Although it is believed that on the Dee
there is a link, this can only highlight the trajectory of change in the population, not quantify total
numbers. Therefore, catch cannot be used as an estimate of population size. Information from two
tributaries supports the view that catch and population size may be related, and that the numbers of
returning adults have declined. Therefore, the evidence suggests that returning numbers are declining,
but this cannot be quantified currently. These reports also suggest that smolt output may be
independent of returning adults, but clearly there is a limit to this and at some point smolt output will
decline if adult returns continue to decline also.
The literature review also conclusively demonstrated that stocking of Atlantic salmon on the Dee will
risk the future of the wild population. The published scientific literature shows that stocking
negatively impacts every life stage of naturally produced salmon stocks, either through direct
competition for resources during freshwater life history stages, reduced marine survival, and via
subtler, but still very important, genetic mechanisms leading to reductions in fitness and an inability to
cope with environmental change at an evolutionary level. Survival rates of stocked fish are held to be
very low, with 0 to 0.1% currently deemed probable. This view is now widely accepted by the
academic, regulatory and management fields and has led to the banning of stocking in Wales. Other
reports show that in rivers were active stocking has taken place, recovery in salmon populations may
be caused by straying from adjacent rivers and not from stocked fish.
Reviewing the electrofishing data proved inclusive in determining the health of salmon juvenile
populations. This is because the variety of drivers for data collection leads to incomplete coverage
across the entire catchment, and the datasets are not long enough to paint an accurate picture over a
number of years. The main concerns on the data collection regime are being addressed by the
Catchment Health Monitoring programme; however, this does not represent a mature dataset
sufficient to support a stocking application and more time is needed. Subsequently the report was
unable to determine an appropriate life history stage to stock, and a precautionary approach via
River Forth Fisheries Trust: River Dee Hatchery Assessment
iv
stocking of smolts was proposed. This will limit the residence time of stocked fish in the river and
their impact on the wild produced fish. However, by maximising a return the genetic risk will become
greater.
Economically, the costs of building and running a hatchery to produce smolts are exceedingly high. To
support a rod catch of 1000 fish, start-up costs of approximately £4 million with annual running costs
of half a million pounds should be expected with no guarantee of any fish produced returning to the
rod. These values were based on an economically realistic value of 4.5% return of produced smolts
representing the central point in the range of stocked smolt survival rates published in the relevant
literature. From a biological perspective this is very high and is unlikely to be attained in practice. It
was used solely to produce a believable economic appraisal with the cost of producing 1000 returned
rod caught salmon at 0.1% survival running into exceedingly high levels.
Values for producing 100 rod caught fish were excluded as this level of increase was deemed
redundant, while the broodstock requirements to produced 5000 rod caught fish could be considered
difficult to support.
Based on the accepted scientific advice, stocking of Atlantic salmon may risk the Condition Status of
the Dee SAC. Any regulatory approval for stocking will need the stocking license applicant to
demonstrate the cause of any decline in salmon population and clearly demonstrate that stocking will
arrest and reverse this decline while maintaining the genetic integrity of the wild stock. These data do
not currently exist to support any of these requirements. The applicant will also need to show that all
other avenues to increase salmon populations on the Dee have been exhausted. Habitat access and
quality are still seriously impacted by landscape and in-channel pressures on the Dee, and improving
the production and survival of wild fish through returning the river to a more natural state should be
prioritised prior to any stocking application.
Other important legal considerations are relevant. Although the Dee Board currently can grant a
license to stock salmon, the license to remove broodstock is not granted by the Board and this will
require the license applicant to satisfy Marine Scotland Science concerns. Ultimately however, due to
the upcoming changes in the management of salmon in Scotland, it is very possible that the new
Fishery Management Organisation on the Dee will not be able to license any aspect of stocking, and
any decision on stocking should wait until the distribution of duties between local and the national
FMO have been clarified.
All strands of investigation show that stocking of the Dee is not appropriate at this stage. Stocking will
risk the biological integrity of the wild population, is enormously expensive and there is no guarantee
of a license. Most importantly, there is also no guarantee that any fish will return. Approaches aiming
to improve the quality and quantity of habitats in the river should be prioritised over stocking.
Although stocking of salmon is carried out in other rivers in Scotland, it has been shown to be
ineffective, a measure of ultimate last resort in rivers which could not support a functioning fishery, or
subject the claims which are unverified and not accepted by the wider scientific audience. Even in
areas where stocking is perceived as successful, such as the Ranga, the Conon or the Bush, the level of
difference between these rivers and the Dee in terms of size, structure and impacts are of such
significance that they cannot be used as suitable comparisons.
River Forth Fisheries Trust: River Dee Hatchery Assessment
v
EXECUTIVE SUMMARY .................................................................................................................. III
1. INTRODUCTION ...................................................................................................................... 1
1.1. CONTEXT ...................................................................................................................................... 1
1.2. THE RIVER DEE .............................................................................................................................. 2
1.3. THE DEE FISHERY ........................................................................................................................... 3
1.4. AIMS AND OBJECTIVES .................................................................................................................... 7
2. BIOLOGICAL ASSESSMENT OF STOCKING ON THE DEE ............................................................ 10
2.1. REVIEW OF SCIENCE AND CURRENT THINKING .................................................................................... 10
2.2. ASSESSMENT OF FISH AND HABITAT DATA ......................................................................................... 12
2.3. CONCLUSION ............................................................................................................................... 14
3. THE COST OF STOCKING ........................................................................................................ 16
3.1. HOW MANY FISH? ........................................................................................................................ 16
3.2. THE HATCHERY COST ..................................................................................................................... 20
3.3. COST BENEFIT ............................................................................................................................. 20
3.4. TIMELINE .................................................................................................................................... 21
3.5. CONCLUSION ............................................................................................................................... 21
4. LEGAL ................................................................................................................................... 24
4.1. PROTECTED SITES ......................................................................................................................... 24
4.2. LEGAL OBLIGATIONS OF THE BOARD: MANAGEMENT OF SALMON .............................................................. 31
4.3. LEGAL OBLIGATIONS OF THE BOARD: ALTERNATIVES TO STOCKING ............................................................. 32
5. DISCUSSION .......................................................................................................................... 40
5.1. WHAT NEXT? .............................................................................................................................. 44
5.2. CONCLUSIONS ............................................................................................................................. 44
6. REFERENCES ......................................................................................................................... 46
ACKNOWLEDGEMENTS ................................................................................................................. 53
1. Introduction
1.1. Context Over recent years the River Dee has experienced a reduction in rod catches of Atlantic salmon (Salmo
salar). These reductions have been sufficient to cause a decline in the five-year average, indicating a
decline in the number of fish caught. There is a now concern that extreme events, such as Storm Frank
in 2015/2016, may exacerbate the declines in catches and perhaps impact the overall population.
Therefore, as one of a number of possible mitigation measures, the Dee District Fishery Board have
commissioned an assessment into the potential for a hatchery to provide a biologically sound,
economically viable and technically feasible solution to boost catch numbers. Stocking of Atlantic
salmon is not currently undertaken on the Dee.
Salmon stocks are subject to naturally driven variability, both cyclical (which can operate on annual,
inter-annual, decadal and multi-decadal cycles) and random (e.g. an immediate response to major
flood event) (Friedland et al., 2003). Therefore, although at a given point in time any measure of
population may demonstrate a decline (or rise) in numbers when compared with a previous year;
when compared with population counts over an extended period of time these declines may appear
quite normal. Therefore, it is possible that the Dee stock may be undertaking a downward trajectory
as part of a natural cycle.
A key issue lies in determining the population size of salmon in the river and the cyclicality of the
stock. Catches are often used as proxy indicators of fish population size, usually because they are the
only long-term data to hand. Unfortunately catch is very often a poor indicator of population size
(Hendry et al., 2007). Therefore, at best they can show us that the Dee population is changing at an
unspecific rate; at worst they present an inaccurate picture. The range of factors which influence
whether an individual angler catches a fish during a fishing trip are substantial (including experience,
fitness, weather, location, fish behaviour, fish cover, water temperature, migration stage and
biological condition of the fish). An angler must be on a river to fish and often the perceived poor
quality of a fishery relating to a bad year (or number of years) may result in a reduced fishing effort
with subsequent decline in catches. All of these factors limit the ability of exploitation data to
determine target population size.
The perceived benefits of stocking salmon are numerous. At the simplest level it will increase the
number of juvenile fish in the river and by extension should increase the number of returning fish and
subsequent catch. By boosting the number of returning adults, a fundamental tenet of stocking is that
it will also boost salmon spawning with increased natural productivity as a result of increased returns.
Stocking has historically been seen as intrinsic to managing fisheries, and potentially integral to
creating a fishery on the Ranga in Iceland, the Bush, Burrishoole and Delphi in Ireland, and the Conon
and Carron in Scotland. These programmes have been driven by a mix of fishery creation, fishery
improvement, and salmon conservation. Two distinct forms of stocking are now considered,
conservation stocking to restore critically endangered populations and fishery enhancement stocking
to improve rod catches (Bacon et al., 2015).
There have been a number of concerns raised about the stocking of salmonids to boost populations
for the purposes of exploitation. Studies (e.g., RAFTS, 2014; Young, 2014; and references therein)
have shown stocking may actually be implicated in the decline of salmon returns through factors such
as increasing stock susceptibility to climate change (Jonsson et al., 2003), increased rates of straying
contaminating the gene pool of salmon from neighbouring systems (Isaksson et al., 1997; McGinnity et
al., 2003) and changing key migration timings.
River Forth Fisheries Trust: River Dee Hatchery Assessment
2
Artificial stocking can reduce the fitness of entire salmon populations through both inbreeding and
outbreeding depression (outbreeding depression – this is where a parent can reduce the fitness of the
next generation by introducing excessively different genes which are not evolved to the local
environmental conditions). The effects of these events are exacerbated by the larger sizes of stocked
juvenile fish which, due to their low energy, low predation and high calorie rearing environment, can
outcompete naturally produced populations in the freshwater environment.
Although much of the recent focus on stocking has concentrated on the genetic factors, in-river
resource availability is also key because habitats are limited and these limitations have repercussions
for salmon production (Solomon, 1985). While the degree of limitation varies between rivers, all
rivers have a maximum carrying capacity for production (Uusitalo et al., 2005). If a river which is
already operating at carrying capacity were to be stocked, the impact on the overall population may
be a reduction in production due density dependent factors, with the river unable to provide for all
fish (P. McGinnity, pers. comm; Bacon et al., 2015), regardless of the physical condition of the
individual.
At a practical level, economics and legal responsibilities are also a factor when discussing the decision
to stock. Based on the economic value of the Dee (adapted and updated from Radford et al., 2004),
each rod caught salmon may be worth over £2,600 pounds. Therefore, if that fish was stocked and
the production cost of that rod caught fish was also over £2,600 then it may not be economically
viable to run a hatchery. Other work has shown that a returned Atlantic salmon may be worth over
£5,000 (Postle and More, 1996 – over £8,000 in 2016 with inflation) or just over £2,000 on the Tweed
(SVQ, 2015).
Legally, the Scottish Environment Protection Agency (SEPA), Scottish Natural Heritage (SNH), Marine
Scotland Science (MSS), the National Park and the Dee Board will all have input into the decision to
stock and the varying requirements of all these groups need to be considered. This will range from
granting of licenses to statutory consultee to recommended consultee.
1.2. The River Dee The River Dee drains a catchment area of approximately 2100km2, and is one of the largest river
systems in Scotland. It rises in the Cairngorm Mountains and enters the North Sea at Aberdeen.
Although often seen as an iconic ‘wild’ highland river, the Dee runs through significant sections of
highly impacted landscapes, with upland land management for grouse and deer, lowland intensive
agriculture and sections of significant urbanisation all affecting instream habitats, water quality, and
hydrology of the catchment (SEPA, 2015).
Sections of the river and its tributaries are found in the Cairngorms National Park. This is the largest
national park in Scotland and the UK, and was founded in 2002 with expansion into Glenshee and
Perthshire in 2010. The River Dee is also designated a Special Area of Conservation (SAC) with
Atlantic salmon one of the Primary Designating Features. The other designating features include the
freshwater pearl mussel (Margaritifera margaritifera) and the European otter (Lutra lutra), both of
which also rely on Atlantic salmon as host (Hastie & Young, 2003) and prey (Carss et al., 1990)
respectively.
Additional environmental regulation of the Dee catchment is overseen by SEPA. Firstly, a license under
the Controlled Activities Regulation (CAR) Act 2011 may be necessary for any hatchery. Secondly, the
River Forth Fisheries Trust: River Dee Hatchery Assessment
3
Scottish Government has a requirement to improve the Ecological Status of rivers as part of the Water
Framework Directive (WFD – European Commission, 2000) and implemented within the second River
Basin Management Plan (RBMP) cycle. Fish community quality is an aspect of this Status.
Management of the Dee salmon stock is within the remit of the Dee District Salmon Fishery Board (the
DDSFB or the Board), supported by the River Dee Trust (the Trust). The primary purpose of the Dee
Board (under the Salmon and Freshwater Fisheries Consolidation Act 2003 and earlier legislation) and
Trust is to conserve and enhance the fishery on the river, while also protecting habitats for these fish
species which form the basis of the fishery. Therefore, it must explore all avenues for the continued
success of the fishery.
1.3. The Dee fishery The Dee is considered a multi-sea winter (MSW) salmon fishery, although multiple stock components
are present. A time series of catches is presented below in Figure 1.1. This figure shows that both
spring and non-spring catches have decreased since the 1950s. For further information on changes in
run timings and by month changes see the Dee stock component reviews (Bilsby, 2008; River Dee
Trust, 2013).
Figure 1.1 - Catches on the River Dee (1952-2015, shaded) with five-year averages for each stock
component (lines) for spring and the remainder (rem) of the year
This chart also demonstrates the varying relationship between catch periods on the Dee, with non-
spring catches becoming increasingly dominant recently.
The link between catch and population size for Atlantic salmon is not very strong (Hendry et al., 1997).
Therefore, while it is easy (and very tempting) to look at a catch count and assume it represents a like-
for-like reduction in the total overall population; this is not the currently accepted view. However,
there is a common perception the Dee catch might echo the general trend in population size,
especially when certain datasets from two heavily monitored tributaries are reviewed
Figure 1.2 charts the relationship between spring rod catch on the Dee and the number of females
recorded entering the Baddoch and Girnock, two tributaries, while Figure 1.3 provides a direct
0
2000
4000
6000
8000
10000
12000
14000
1952 1962 1972 1982 1992 2002 2012
Cat
ch
1st June to end of season
Spring Feb-May
5YA Rem
5YA Feb-May
River Forth Fisheries Trust: River Dee Hatchery Assessment
4
comparison. Both these figures suggest that periods of high catches may correspond with periods of
high female counts1 (but, when extrapolating the counts of fish from two small tributaries to compare
with an entire fishery, care must be taken).
Figure 1.2 - Time series of Dee spring rod catch and trapped females from the Girnock and Baddoch
Figure 1.3 - Relationship between spring adult catches (Dee catchment) and trapped females in two
Dee tributaries (Baddoch and Girnock)
So, while there may be potential in determining the general patterns of population increases and
declines using catch data on the Dee, there is still no way of using these data to determine the
absolute number of fish returning2.
1 Due to a statistical anomaly, direct comparison may not be appropriate. Therefore a more mathematically
correct (transformed) graph representing the comparison between Dee rod catch and female fish on the two tributaries is presented in Appendix A. Both charts reflect the explanation in the main text. 2 This does not suggest that local knowledge is not a key factor in understanding a fishery. Very often
proprietors, ghillies and anglers are the first identifiers of problems, and previous discussions between the author and salmonid fishery managers has identified the urgent need to formally record this knowledge base as people retire or move on.
0
50
100
150
200
0
2000
4000
6000
8000
10000
12000
1950 1960 1970 1980 1990 2000 2010
Trap
co
un
t
Cat
ch
Spring rod catch
Girnock females
Baddoch females
1
21
41
61
81
101
121
141
161
181
0 1000 2000 3000 4000 5000 6000 7000 8000
Trap
co
un
ts
Spring catch
Girnock females
Baddoch females
River Forth Fisheries Trust: River Dee Hatchery Assessment
5
The most common option of establishing returning adult numbers is direct counting of migrating
adults using fixed fish counters, as introduced above in Figures 1.2 and 1.3. There are two of these on
the Dee catchment (Girnock Burn – Figure 1.4a) and Baddoch Burn – Figure 1.4b) and they are
commonly used elsewhere (for a full, but perhaps slightly out of date, inventory in Scotland see
Simpson, 2003). These structures require some way to ‘funnel’ fish into a space where they can be
recorded by a number of means (e.g. resistivity or cameras) so, quite often they are placed on existing
structures such as weirs and fish passes which provide a suitable platform to force fish alongside the
recording apparatus. However, weirs and dams have negative implications for salmon populations,
and even fish passes may not be considered fish friendly. (Despite perceived efficacy, there is a lack of
robust and quantified data on fish passage and data suggests passage is much less than expected –
Noonan et al., 2012). These structures also significantly impact the cleaning, renewal and creation of
new habitats, especially spawning habitats.
A B
Fig. 1.4 - Fish counting devices on the Girnock burn (A) and Baddoch Burn (B)
The location of the counter is also critical. To successfully count all the adult fish entering a system
the counter must be located close to the tidal limit at the rivers’ confluence with the estuarine
environment, or at least below the first major tributary. This is not available on the Dee, and
considering uncertainty regarding fish passage beyond weirs and their severe impact on river habitat
and communities (Poulet, 2007), building a structure simply to count fish will certainly have a negative
impact on the Dee and should not be considered; only an existing structure or alternative method
should be modified. So while the counter information on the Baddoch and Girnock are useful, like
catch data they should be considered only as potential indicators of population health and general
population3.
The counting of juvenile fish is another approach to establishing the effective population size; either
by direct surveying using electrofishing or by counting juvenile fish as they migrate to the sea as
smolts. The assessment of population size using juvenile electrofishing can be an effective
management tool, however accurate surveying requires a significant allocation of resources before a
3 During discussions with H. Moir he stated that although the dataset on the Girnock is an excellent resource,
the variation in river types found in the Dee tributaries might not make it the most appropriate dataset from which to draw inferences on the Dee catchment as a whole.
River Forth Fisheries Trust: River Dee Hatchery Assessment
6
robust long-term picture of population health across an entire catchment is captured and, in reality,
this may not be possible on the Dee4. The Dee electrofishing data will be explored in Section 2.
The trapping of seaward migrating juvenile salmon on the Dee has many constraints in common with
trapping migrating adults. Similarly, while it might be considered a suitable surveillance method, it
might not be the best indicator of total population size. However, once again it can provide a general
pattern of river health. Smolt data from the Girnock and Baddoch traps is presented below in Figure
1.5.
Figure 1.5 - Smolt counts from the Girnock and Baddoch Burns
These data show that, although there is a lot of variation in smolt counts when looked at on a year by
year basis (interannual variation), over the longer term the outputs are relatively stable for both
streams5. This corresponds with a more formal analysis carried out by Gurney et al. (2008).
So, it is very difficult to pull together a robust and unequivocal snapshot of the current state of the
Dee Atlantic salmon stock and how the size of this stock definitively relates to rod catch. Based on the
information gathered from catch returns, there has been a general decline in the number of adult fish
caught. A decline is also evident from counter data gathered from two tributaries. Although smolt
output is clearly variable (for where we have information), there may not be an associated long term
decline in the number of smolts migrating through the river. This information comes with many
caveats, not least of all the fact these data only come from two small tributaries, but it does raise the
question on whether the Dee (or areas of the Dee) may be approaching carrying capacity, through
some mechanism controlling one or many juvenile stages, for smolt output.
4 Various methods of electrofishing can be used. The most effective and data rich involves netting off a section
of river and electrofishing repeatedly until the majority of fish have been captured. While this is possible on small stream and burns less than 10m in width, surveying the main stem of the Dee in this fashion is often not possible. 5 See Appendix A – Figure A2 - for a more rigorous averaging method which demonstrates the stability in the
smolt output.
0
500
1,000
1,500
2,000
2,500
3,000
3,500
4,000
4,500
1967 1972 1977 1982 1987 1992 1997 2002 2007 2012
Smo
lts
Girnock Baddoch
5YA Girnock 5YA Baddoch
River Forth Fisheries Trust: River Dee Hatchery Assessment
7
1.4. Aims and objectives The objective of this project is to answer the following question: Is a hatchery a practical, cost effective, sustainable and beneficial way of boosting rod catches of adult Atlantic salmon in the Dee? This report aims to present an independent review of the requirement for a hatchery and stocking
policy on the Dee based on all the factors raised above. This review will be impartial with no bias
towards either outcome. It will produce outputs that illustrate how certain decisions were arrived at
in a clear and concise fashion and will include consultation with the Dee Board and Trust at all critical
stages. It will review the applicability of stocking to three rod catch levels of improvement; 100, 1000
and 5000 rod caught fish per annum.
Specifically, the following tasks will be undertaken:
Biology
• Undertake a high level literature review and incorporate this information with previous reviews to assess the applicability of such information to the Dee fishery;
• Review catch and juvenile density records to assess the long term stability of the Dee salmon population;
• Review the available in channel habitat to assess its use by salmon; and • Identify the most appropriate life history stage to stock.
Practicalities/Economics
Develop an indicative cost to develop a hatchery and yearly running costs to stock the identified stage; and
Undertake a cost benefit analysis.
Legal • The Board has an obligation to manage and ensure the continuity of the Dee fish stock. If a hatchery
is not deemed necessary, then how can the board discharge these duties? • If a hatchery is deemed necessary, then how can those operations remain consistent with regulatory
agency requirements, especially with regard to the Dee’s SAC status and obligations under Natura? Although not part of the original proposal, it was suggested that discussions with certain people could help support the project. The project team approached a number of individuals and the full list of people with whom discussions were held is presented below. Not all individuals will be credited with comment in this report.
Colin Bean, Scottish Natural Heritage;
Will Boyd-Wallis, Cairngorm National Park;
Gavin Clark, SNH;
Alastair Duguid, SEPA;
Richard Kennedy, River Bush Salmon Station;
Robert Kindness, River Carron;
Ian Malcolm, Marine Scotland Science;
Hamish Moir, cbec eco engineering UK;
Simon McKelvey, Conon Fishery Board;
River Forth Fisheries Trust: River Dee Hatchery Assessment
8
Phil McGinnity, University College Cork;
Debbie Parke, Nith Catchment Fishery Trust;
Brian Shaw, Spey Foundation;
Eric Verspoor, University of Highlands and Islands; and
John Webb. A series of queries were supplied to a number of nominated stakeholders for short, structured responses. This was undertaken to gauge opinion and perception on the Dee stock and the state of the fishery and was not intended as a full and formal survey. Individuals contacted for comment were proposed by the Dee Trust and all respondents replied. Titles are based on the individual’s position at the time of receipt of the queries, prior to the September 2016 Board meeting. The queries posed were:
1. In your opinion, do you feel that the Dee fishery is in a chronically bad state, or is it just going through a ‘bad period’ that you expect the river to recover from?
2. If you feel fishing on the river is undergoing a long term decline, what do you think are the three main causes of this decline?
3. Although stocking has been commonly used in the past as a management tool for
Atlantic salmon fisheries, there appears to be a growing body of scientific evidence that stocking can do long term harm to salmon populations. How aware are you of this information, and how do you feel this information fits to the Dee?
4. Has this scientific information been communicated to you in an accessible and easily
understood fashion?
5. Has your estate or fishery ever stocked salmon, and if so when was the last time you stocked?
6. In your opinion, how effective was this stocking, if it occurred?
7. How effective do you believe stocking on other Scottish rivers is?
8. If stocking was to be allowed on the Dee, how effective do you think it would be?
9. If stocking was not to be permitted on the Dee, which of these options would you
support to increase catches and/or salmon population size: a. subsidised angling to increase fishing effort b. complete removal of weirs to ensure unhindered passage c. reforesting the catchment to improve nutrient retention in the headwaters and
increase habitat diversity in the main channel d. river restoration projects to remove embankments and re-meander straightened
sections e. funding research into improving marine return rates f. Stocking is the only option we should consider.
River Forth Fisheries Trust: River Dee Hatchery Assessment
9
10. The decision to stock the Dee may not only be made by the Dee Board/Trust, and other agencies such as Scottish Natural Heritage/Cairngorms National Park and Marine Scotland Science may object to it. If this was to occur how would feel about this
11. Assuming a new building is required to house a hatchery to support stocking, at which
point do you think the build cost is too expensive: a. £100,000 b. £500,000 c. £1,000,000 d. £2,500,000 e. £5,000,000 f. £10,000,000 g. Cost is irrelevant?
12. At which point do you think the yearly costs of running a hatchery become too
expensive: a. £50,000 b. £100,000 c. £250,000 d. £500,000 e. Cost is irrelevant?
13. Would you be prepared to increase your levy contribution to fund stocking operations?
14. As a percentage, with 100% being no vacancies on your fishery, how many of your available fishing days are being used?
The responders are listed below:
Ghillies Association; Aberdeen District Angling Association Park House; Dinnet and Kinord Estate; Kincardine Estate; Dess Estate; Lower Blackhall & Kinneskie and Invery; and
Anon.
Responses are given in full in Appendix F, and referred to throughout. Not all queries were answered by all respondents.
River Forth Fisheries Trust: River Dee Hatchery Assessment
10
2. Biological assessment of stocking on the Dee
2.1. Review of science and current thinking Throughout its natural range, both within and between catchments, Atlantic salmon are commonly
found as distinct populations adapted to local catchments with little interbreeding among them. These
genetic differences are due to a combination of evolutionary factors (e.g. mutation and natural
selection) over time, and they are maintained by both a salmon’s homing behaviour for breeding and
local environmental pressures. The quantification and conservation of the levels and patterns of
population genetic diversity in Atlantic salmon is now recognised to be of fundamental importance for
the long-term sustainability of local populations. It is this natural genetic diversity that enables
populations to respond (i.e. to adapt) to the continuously changing river environment and, hence, to
persist over time.
As a consequence of human-mediated activities (e.g. urban growth, farming activities, loss and/or
fragmentation of suitable spawning and nursery habitats, global warming, overfishing); salmon have
declined throughout most of its natural range in the North Atlantic region (WWF, 2001). Decline in
wild stocks has led to widespread releases of artificially spawned and reared Atlantic salmon in an
attempt to supplement natural production and or to mitigate real or perceived wild population
declines. In particular, hatchery releases are undertaken in an attempt to ‘enhance’ fishing for
recreational angling and tourism, which are economically important in most countries where Atlantic
salmon is still found (McGinnity et al., 2004).
From a population restoration, supplementation and/or conservation perspective(s), salmon stocking
involves the intentional release of hatchery individuals into the wild to boost natural production. The
main rationale for stocking is that survival from ova to fry or parr in a well-protected hatchery
environment is thought to greatly exceed that in the wild and, potentially, allows for a “cost effective”
management approach to enhance natural production. Notwithstanding the presumed benefits,
however, stocking practices have been increasingly and widely challenged/criticized for their low
efficiency in preventing wild populations declines (limited usefulness and poor economic value) and
their negative impacts on the fitness of individual populations, which is intrinsically linked to long-term
population sustainability (Aprahamian et al., 2003; McGinnity et al., 2003, Araki et al., 2008; Fraser
2008; McClure et al., 2008).
The negative impacts of interbreeding between non-native and wild salmon are well documented in
the scientific literature (e.g. Fleming and Einum 1997; McGinnity et al., 2003; Araki et al., 2007, 2009;
Hutchings and Fraser 2008; McLean et al., 2008; Thériault et al., 2010, 2011). The unambiguous
consensus is that changes in the genetic composition of wild populations caused by stocking has the
potential to reduce the evolutionary capacity of populations to respond to changing environments
and, hence, to compromise the long-term viability of local stocks. Indeed, in more extreme cases
involving continuous interbreeding between wild and non-native salmon, stocking may potentially
lead to localised population extinctions (McGinnity et al., 2003).
In an attempt to overcome potential problems related to interbreeding with non-local individuals, a
common idea behind modern supplementation/restoration stocking programs is to establish hatchery
fish from local broodstock (i.e. local adult females and males). The hatchery fish are reared fish in
protected environments, and subsequently released at any stage of their life cycle (e.g. eggs, fry, parr,
River Forth Fisheries Trust: River Dee Hatchery Assessment
11
and smolts) in to the wild. If broodstock fish are sampled on a yearly basis, it was historically thought
that their offspring will closely reflect the natural spawning behaviour of wild adult salmon, hence
diminishing the likelihood of impacting local wild populations. One relevant caveat of this approach,
however, is that forced coupling may have genetic implications for the next generation (Neff et al.,
2008; Anderson et al., 2012). Another important factor to consider in the hatchery environment is that
mortality levels and hence selection pressures are very different from the wild. It is often the case that
selection will decrease the fitness of hatchery fish once they are released into the wild (Araki et al.,
2007).
There are many obvious differences between hatchery and wild environments including:
Fish density;
Changes in habitats;
Interactions with other salmon and other fish species such as trout; and
Food availability.
All of these are likely to lead to genetic differences linked to important traits (e.g. growth rate, fish
morphology, and behaviour) between wild and hatchery fish (Metcalfe et al. 2003; Hawkins et al.
2008). For instance, it is well known that hatchery-reared fish generally have higher growth rates than
their wild equivalents (Berejikian et al., 2000). These differences may influence the resulting life-
history of stocked fish, in particular a reduction of the average number of winters they spent at sea in
comparison to wild-born fish (Jonsson and Jonsson 2006; Kallio-Nyberg et al., 2011). As body size (and
fertility) is affected by time spent at sea, the reproductive success of hatchery born/reared fish in the
wild may be affected (Thériault et al., 2010, 2011).
Leading from this last argument, it is important to realise that the major assumption of supplemental
stocking programs is that the hatchery-born fish will display a similar level of reproductive success in
the natural environment in comparison to their wild born counterparts. From the arguments outlined
above, this key assumption may not be the case. Indeed, a number of recent scientific studies,
capitalising on modern genetic techniques, are challenging this. For instance, studies on the chinook
salmon (Oncorhynchus tshawytscha), steelhead (O. mykiss) and coho salmon (O. kisutch), have clearly
demonstrated that a single generation of hatchery rearing of local fish was sufficient to decrease the
reproductive success of individuals once they were released into the wild (Araki et al., 2007,
Williamson et al., 2010, Thériault et al., 2011).
Similar results have also been reported for Atlantic salmon. Milot et al., (2012) assessed the
reproductive success of wild and hatchery derived Atlantic salmon in a small river in Québec. The
stocking strategy involved yearly release of fish following a single generation of captive breeding.
Among the adults returning to the river to breed, between 11% and 41% each year were born in the
hatchery. The authors found out that the relative reproductive success (RSS) of the fish born in the
hatchery was approximately only half of wild-born fish. The study concludes that being reared in a
hatchery environment altered life-history strategies of hatchery born fish and negatively affected their
future reproductive success.
In summary, current scientific evidence suggests that while stocking is a potentially feasible strategy
for supplementing natural Atlantic salmon stocks, its use should be carefully considered. Indeed, the
argument is that hatchery intervention should be the last option, and it should only be carried out as
River Forth Fisheries Trust: River Dee Hatchery Assessment
12
part of comprehensive fishery management and biodiversity conservation plans and not as an
independent approach. Thus, in the case of scientifically proven population declines, the first step, and
prior to any management action, is to establish the reasons for the decline (e.g. limited number of
spawners, lack of suitable spawning habitats, poor survival at different life-history stages, food
limitation, habitat degradation, barriers preventing movement, water chemistry, temperature, etc.) in
addition to the level of decline. This is consistent with the likely approach required by Scottish
regulators.
Management actions should be based on evidence-based facts related to a particular environment
and/or population. Thus, a given management strategy that may potentially work for one particular
situation, may not work in other. For instance, if the low number of fish in a given environment is
linked to food limitation and or suitable habitat (carrying capacity), stocking will result into increased
competition, which, because of density dependent survival, may lead to even lower numbers of fish in
the long run.
2.2. Assessment of fish and habitat data This section provides a summary of the complete review presented in Appendix B.
A stock assessment review carried out by the Dee Trust in 2015 presented a classification of site
quality based on a comparison of observed fish densities (O) against an expected density (E) calculated
using Millar 2015. This report did not use this approach as it not only had to develop an indication of
the catchment scale quality of the Dee juvenile stock, but also the quality of these data used to
understand the long term stability and patterns of juvenile production and survival on the Dee with
specific reference to the potential implementation of stocking. Therefore, this data review has a
different remit to the Dee stock assessment which was tasked with understanding quality based on
data benchmarked against a national model at the site scale.
Available data
Four main sets of data were provided to assess the distribution, density and habitat use of juvenile
Atlantic salmon on the Dee. These were three electrofishing data sets from 2011 to 2015, fully
quantitative, semi quantitative and timed, and the habitat walkover. Table 2.1 below shows the total
size of these datasets ultimately used for the juvenile assessment. Figure 2.1 shows the location of all
electrofishing data set against the sections of channel surveyed as part of the habitat walkover.
Table 2.1 - Electrofishing data description
Type 2011 2012 2013 2014 2015 Replicate
sites Habitat
data 2016
Timed 96 28 56 73 78 66 53 N/A*
SemiQ 15 30 34 49 48 49 22 11*
FullyQ 36 30 38 22 25 31 8 6*
Total 147 88 128 144 139 146 83 17*
* Incomplete data set
River Forth Fisheries Trust: River Dee Hatchery Assessment
13
Figure 2.1 – Distribution of electrofishing sites
As Table 2.1 shows, not all data could be used in the juvenile assessment. Only sites which had
repeated surveys could be used because the risk of an unusually low or unusually high event biasing
the assessment could not be controlled if all data were used. Secondly only those sites which had
corresponding habitat data could be used as part of the assessment included investigating the
relationship between the biology data and habitat. Some data from 2016 were also provided to assess
any impacts from the major flooding over the winter.
Assessment The datasets were reviewed based on the median density at the site, the level of yearly change in the
densities and the related habitat. As three separate surveys methods were used and these data might
not be confidently compared across the three methods, individual assessments were carried out for
each method. The local habitat, based on the location of the survey, was identified to establish the
site conditions which could (or should) affect the density of juveniles and the life history stages
present.
The relationship between 0+ fry and 1+ parr was also investigated, to help understand the patterns of
survival in the Dee. This approach also can provide some information on the potential impacts of the
severe winter flooding, and a specific comparison between the density of parr and fry between 2015
and 2016 was carried out.
Results The distribution of the sites shows that significant sections of the catchment remain without robust
data with which to assess the juvenile population. There is clustering of surveys in certain tributaries
and sections of the main stem, but there is no long-term, consistent and repeated survey approach
River Forth Fisheries Trust: River Dee Hatchery Assessment
14
across the Dee. This means that some locations, such as the Gairn and Clunie have a good level of
survey coverage, while others (especially the main channel) have fairly limited coverage.
The results show that the only location with a good characterisation of the juvenile communities is the
Gairn, which appears to have a fairly stable juvenile population. Densities appear to be at medium or
sometimes high levels relative to the rest of the catchment. However, due to the variable nature of
the other sites, it is not possible to draw robust conclusions elsewhere. The habitat information shows
that there is very little fry and spawning habitat associated with the survey locations, and parr habitat
dominates.
When the survival of juveniles from 0+ and 1+ was reviewed there appeared to be a consistant rate
throughout the years with available data, and this was consistent across survey methods. A direct
comparison of densities showed that there may be an upper density of parr which is independent of
the number of fry from the preceding year. However this is a very tentative observation and should
only be used to justify further investigation into the relationship between the two life history stages.
A very small of amount of data (approximately ten sites each from semi quantitative and fully
quantitative surveys) was available from 2016 to compare with the historic dataset (2011-2015). The
overwinter survival rate (2016) was consistent with the previous winters, and there appeared to be no
major declines in fry or parr observed between 2015 and 2016. These observations are very heavily
caveated as the number of available surveys is too low to draw confident conclusions.
2.3. Conclusion The biological review in Section 2.1 conclusively argues that stocking of salmon should be carefully
considered due to the scale of potential negative impacts on the naturally produced Dee salmon stock.
There is unambiguous evidence that stocking of salmon will reduce fitness of populations, exacerbate
the impact of declining marine survival, and risk not only salmon from the Dee but potentially salmon
from other catchments. It also makes the very important point that it should not be considered until
the cause of the decline in salmon catch has been understood and quantified.
The collection of electrofishing data on the Dee is often project specific. This will allow the datasets to
be used, in time, to establish the levels of success associated with a range of works and help with
replicating any major successes. But the clustering of the locations makes it difficult to draw a
catchment scale picture. The recently activated Catchment Health Monitoring programme should
eventually deliver the detail required, but currently the electrofishing data does not lend itself to a
catchment scale and robust juvenile assessment, and therefore cannot be used to determine the need
for stocking, or to identify the most appropriate stage to stock.
This also affects any attempt to link habitat use and juvenile fish numbers (i.e. the carrying capacity) at
the local level and catchment level. There appears to be shortage of spawning and fry habitats on the
river based on a visual review of habitats from the walkover and the available habitats adjacent to the
electrofishing survey sites, however when we look at the relationship between fry and parr there
might be a (slight) suggestion that parr outputs are limited. However, it should be strongly caveated
that this cannot be more than a suggestion from a very constrained dataset. Based on the walkover,
River Forth Fisheries Trust: River Dee Hatchery Assessment
15
summer parr habitat is certainly not limiting, so we could perhaps focus on refuge and overwintering
habitat as one of many potential limiting factors6.
The electrofishing and habitat data presented within this section is unable to identify conclusively any
bottlenecks in productivity of juvenile salmonids. This is not because they do not exist; the evidence
from the Girnock and Baddoch burns shows that smolt output is fairly stable irrespective of the
number of females returning. Unfortunately, with regards to a catchment level assessment, the spatial
extent(s), lack of data from the mainstem and the varying methods employed to collect that data all
coalesce to reduce the effectiveness of the dataset(s). This is in agreement with the Dee 2015 stock
review which, although being able to apply a more definite quality score to the electrofishing
densities, stated that significant areas of the catchment were not categorised and the dataset used
may not have allowed a full catchment scale assessment to be produced. The chart showing the
relationship between 0+ and 1+ salmon hints at parr output being limited, but this interpretation is
suggestive and subjective at this stage and should only be used as the basis for further investigation.
Therefore, based on the only robust available data, smolt output is limited (on the Girnock and
Baddoch burns) and in the absence of any other information a conservative approach to stocking
should be taken. The stocking of ova, fry or parr might not be effective and may unbalance density
dependence mechanisms controlling production (P. McGinnity, pers. comm; Bacon et al., 2015),
especially if juveniles are stocked into a potentially resource limited environment. These limitations
can be overcome by stocking smolts. This is because the smolt stage will not take up any habitat in
the river beyond migrating through them, and the risk of overwhelming habitats is not high. From a
fishery perspective stocking of smolts is more likely to result in a returned salmon; however, from a
population perspective this may increase the genetic risk (P. McGinnity, pers. comm).
Based on very initial comparisons, there is no evidence that Storm Frank has reduced fry and parr
densities in those sites with suitable data. Realistically, this dataset is not sufficient either in terms of
the coverage of sites or the replication of surveys to determine any impact on the salmon population
and it is not appropriate to draw any conclusions regarding that event at this time.
6 The mobility of fish does mean that they can travel significant distances when seeking overwintering habitat.
This may mean that overwintering habitat is one of many potential controls.
River Forth Fisheries Trust: River Dee Hatchery Assessment
16
3. The cost of stocking Although stocking operations have previously been fairly ad hoc operations with small hatcheries run
by a limited number of dedicated people or even individuals (for example the Carron), in practicality it
is now a very large undertaking. For example, annual running costs on the Spey are between £60,000
and £70,000, and this has been scaled back significantly by the Spey Fishery Board in recent years (B.
Shaw, pers. comm). This cost allows for the production of approximately 250,000 autumn parr, with
attributed adult catch to the stocked juveniles varying from 0 to 1.8% of the annual rod catch over the
course of five years (Coulson et al 2012; Stewart et al, 2015). (However the study period covered a
period when stocking levels were much higher, with up to 2.2 million fed fry released instead of
autumn parr.)
Annual running costs are only one aspect of the monetary cost, and clearly the costs associated with
the Spey are not sufficient boost rod catches. Therefore we cannot assume the costs and magnitude
of the Spey stocking programme will be sufficient to boost rod catches on the Dee. (For comparison,
this study has been asked to investigate the requirements to boost the Dee by 100, 1000 and 5,000
fish – the Spey returned fractions of these values).
The small private hatcheries mothballed currently on the Dee are potentially no longer relevant to
modern stocking operations. This is because:
The previously perceived effectiveness is now seen as questionable;
Changes to the regulatory and legislative environment (such as the creation of the National
Park in 2003 and the SAC in 2005) mean it is no longer within the remit of the Board alone to
action stocking; and
Changes to best practice, such as holding suitable numbers of broodstock, mean they could be
of insufficient size.
The planning permission process for building the hatchery is likely to be extremely onerous, while CAR
licensing means significant assessment of the water needs of any hatchery will be required. There will
also need to be other investigatory procedures (this will be explored further in Section 4) relating to
the SAC. Other important considerations such as biosecurity are now strictly controlled, while it will
no longer be acceptable on welfare grounds to accept the sporadic losses of year classes which often
occurred. Regulation is now supported by accepted standards such as the RSPCA welfare standard for
farming of salmon which would include hatchery operations for stocking (RSPCA 2015).
Consequently, the hatchery will need to be a modern, well designed and efficient building and
operation developed to strict international standards. This project team has included hatchery
designers from Stirling University and Aqua EcoSystems BV to assist in planning and costing the
hatchery, while determination of capital costs (land, build costs etc.) have been developed by a
chartered Quantity Surveyor.
3.1. How many fish? The scope of the project required the team to investigate the size of hatchery needed to produce 100,
1000, and 5000 rod caught fish. To do this it first needed to answer two questions – what is the
angling exploitation rate, and how many fish will return as adults.
River Forth Fisheries Trust: River Dee Hatchery Assessment
17
Exploitation rates are difficult to assess, mostly because we simply do not have a very strong
understanding of the number of fish which return to our rivers (see Section 1). Even in those systems
where efficient counter operations are implemented, factors such as predation and poaching reduce
the available stock for exploitation. Underreporting of catches is also a potential issue, while catch
and release can blur the picture further (see Smith et al., 2014 for an assessment of the potential
effect of catch and release). An additional factor is the potential differentiation between rod capture
rates for separate stock components.
Exploitation rates of UK salmon range from 5% to 35 % (Thorley et al., 2007). Of that reported range,
only one figure was developed from the Dee - 15% (Webb 1998). Other Scottish figures included 5% –
30% from the Spey (Laughton, 1991). Unfortunately, as with many Scottish studies relating to this
project, the number of samples used to develop these rates is very low. To reflect this uncertainty we
chose to develop cases for three categories of angling exploitation (5, 15 and 30%) to represent low,
middle and high rates within the published range.
As suggested in Section 2, the most appropriate phase for fishery enhancement is to stock is smolts.
The ranching or stocking of smolts is a common occurrence, being implemented on the Ranga in
Iceland (Johannsson et al., 1996); the Rivers Bush (Kennedy & Crozier, 1982); Burrishoole (Mills &
Piggins, 1983); Delphi; Tay, Lochy, Dionard and Conon (Stewart et al., 2015) and Tyne (pre-smolts –
Milner et al., 2004). Somewhat expectedly, there is a wide range of estimates for the survival of that
stocked life history stage to the rod, ranging from no return (Tay - Stewart et al., 2015), to 8.2% return
on the Bush (Kennedy & Crozier, 1993). These values also span over 20 years, during which period
survival at sea has altered markedly. Therefore once again we set up a low, medium and high return
rate for each life history class from Aprahamian et al., (2003 – and references therein), Coulson et al.
(2014) and information from Hawkins (PhD thesis) which ranged from 0.1%7 to 8.2%, and as per the
exploitation rate a low (0.1%) middle (4.5.%) and high (8.2%) were determined. The number of smolts
required for each rod return rate, survival rate and exploitation rate are shown below in Figures 3.1-
3.2.
7 The lowest survival rate is 0, but this cannot be used when estimating the cost of a hatchery.
River Forth Fisheries Trust: River Dee Hatchery Assessment
18
Table 3.1 - Estimated numbers of smolts to produce 100 rod caught salmon for a range of exploitation and survival rates
Exploitation % 5 15 30
Adult return 2000 667 333
Survival Value High Middle Low High Middle Low High Middle Low
Survival % 8.2 4.5 0.1 8.2 4.5 0.1 8.2 4.5 0.1
No. of Smolts 24,390 44,444 2,000,000 8,130 14,815 666,667 4,065 7,407 333,333
Table 3.2 - Estimated numbers of smolts to produce 1000 rod caught salmon for a range of exploitation and survival rates
Exploitation % 5 15 30
Adult return 20000 6667 3333
Survival Value High Middle Low High Middle Low High Middle Low
Survival % 8.2 4.5 0.1 8.2 4.5 0.1 8.2 4.5 0.1
No. of Smolts 243,902 444,444 20,000,000 81,301 148,148 6,666,667 40,650 74,074 3,333,333
Table 3.3 - Estimated numbers of smolts to produce 5000 rod caught salmon for a range of exploitation and survival rates
Exploitation % 5 15 30
Adult return 100000 33333 16667
Survival Value High Middle Low High Middle Low High Middle Low
Survival % 8.2 4.5 0.1 8.2 4.5 0.1 8.2 4.5 0.1
No. of smolts 1,219,512 2,222,222 100,000,000 406,504 740,741 33,333,333 203,252 370,370 16,666,667
River Forth Fisheries Trust: River Dee Hatchery Assessment
19
When deciding on a survival and exploitation rate for this section it was important to focus on the
economic, not the biological, and a strong driver was to produce an economically acceptable outcome.
Therefore we provisionally sense-checked the numbers of smolts required against the cost to produce
one smolt from Stewart et al (2015) of £1.50. This is shown below in Table 3.4.
Table 3.4 - Cost of production based on Stewart et al. (2015)*
Exploitation rate
5 15 30 Catch
Surv
ival
rat
e
0.1% £3,000,000.00 £1,000,000.50 £499,999.50
100 4.5% £666,666.00 £22,222.50 £11,110.50
8.2% £36,585.00 £12,195.00 £6,097.50
0.1% £30,000,000.00 £10,000,005.00 £4,999,995.00
1000 4.5% £6,666,660.00 £222,225.00 £111,105.00
8.2% £365,850.00 £121,950.00 £60,975.00
0.1% £150,000,000.00 £50,000,025.00 £24,999,975.00
5000 4.5% £33,333,300.00 £1,111,125.00 £555,525.00
8.2% £1,829,250.00 £609,750.00 £304,875.00 *Rounding error present
Based on the £1.50 unit price:
Using the 0.1% survival rate; over 666,000 smolts would be needed to produce 100 rod caught
fish at a cost of £1 million;
This rises to £10 million for 1000 additional rod caught fish; and
£50 million for 5000 fish.
To use the 0.01% figure (the lowest non-zero return rate in Stewart et al., 2015) would have
increased those estimated costs by ten.
These figures were deemed to be economically inappropriate.
The 8.2% figure was universally agreed to be too high a survival rate. Subsequently the 4.5% return
rate was used. While the 4.5% value is only representative of a median of survival it is a useful level for
an economic assessment with Table 3.4 demonstrating that the associated costs are potentially in line
with expectation. From the biological perspective the only available Scottish data suggests that
survival may currently be much lower and there should be an acceptance that a return of zero (or
close) rod caught fish is very possible.
With respect to the exploitation rate, the task was made somewhat easier as 15% also corresponded
to a published exploitation rate on the Dee.
Therefore a 4.5% return with 15% exploitation rate was used as the basis for costing a suitable
hatchery. Using these rates, for 100 rod caught fish we will need approximately 14,800 smolts per
annum; for 1000 fish 148,000 smolts, and for 5000 rod caught fish 740,000 smolts.
River Forth Fisheries Trust: River Dee Hatchery Assessment
20
3.2. The hatchery cost The description to plan, design, build and run hatchery are presented in Appendix C. Table 3.4
provides the initial investment cost, the initial investment and running cost per fish, and the
subsequent running cost per fish. Further information on operational costs (opex) is presented in the
Appendix.
Table 3.4 - Costs to build, run, and per fish cost of a Dee fish hatchery
Catch Smolts
Hatchery Per rod catch cost
Area of building
(m2)
Capital budget (incl.
VAT)
Annual costs (excl. VAT)
Capital Life cycle cost
1st lifecycle cost
100 14,815 535 £1,384,216 £353,983 £13,842.16 £10,619.49 £24,461.65
1000 148,148 2,176 £3,666,082 £557,465 £3,666.08 £1,672.40 £5,338.48
5000 740,741 5,608 £7,987,119 £871,452 £1,597.42 £174.29 £1,771.71
The value of running a hatchery to produce an additional 100 rod caught fish is highly debatable (P.
McGinnity, pers. comm.) given the relatively large number of returning fish, and this level of
enhancement is seen as a ‘drop in the ocean’. It will also be shown that this level of rod catch increase
has been potentially provided by the removal of coastal nets on other catchments (See Section 4 –
Legal).
Stocking to the 5,000 level would require the removal of 508 adult fish for broodstock. If these fish
were removed annually by electrofishing, then this would be a loss of over 75 fish to the Dee fishery
based on 15% exploitation. If we assume that the 4.5% survival rate was lower (e.g. 1%), then we
would require 2,286 broodstock fish per annum with an annual food bill of over £200,000 (food bill for
508 broodstock fish is £55,000).
The 1000 level remains. This requires 102 fish for broodstock (£10,000 for feed p.a.), over 2000 square
metres of work space, an initial outlay of over £3 million with annual running costs of over
£500,000. This has been calculated using the middle range values of exploitation 15% and return
survival of 4.5%. If stocked smolt return reduced are lower on the Dee (for example Table 3.2 shows
that if survival is 0.1% then over 600,000 smolts will be needed), then the investment needed to
produce 1000 rod catch will far exceed the costs presented here.
3.3. Cost Benefit The price of a rod caught salmon is not standard across fisheries. The river being fished, the likelihood
of capture, the confidence of the angler and the likelihood of capturing a large MSW fish all have a
bearing on any value estimated.
The economic value is also relative and subject to fluctuation. A salmon caught in a fishery which is
deemed to be underperforming or not performing will have a lower economic value than a salmon
from a fishery which is perceived to be performing to a high standard (e.g., people will pay more to
fish the Findhorn than the Forth). Similarly, if people can fish the Don for a fraction of the price of the
Dee with the same perceived chance of catching a fish, then they are more likely to fish the Don.
Therefore, depending on the returns from a fishery a salmon has a value which is more or less than
River Forth Fisheries Trust: River Dee Hatchery Assessment
21
previously judged. And finally, with conservation plans now in place, an angler may pay a premium to
be allowed to kill a fish. Therefore, the choice of value to use for a cost benefit analysis is subject to
fairly broad range of factors and figures.
During tendering, we proposed the value of a Tweed salmon, £2,055.78 (calculated from the £24
million value of the Tweed to the local economy – SQW, 2015 - divided by the five year average catch
returns). Another value of £5,925 was proposed by Postle and Moore (1996), which would be
currently over £8,000 based on a CPI inflation adjustment. However a value produced by the Dee
Trust, updated from Radford et al. (2004) of £2,670 will be used.
Based on Table 3.4 (above) and the Dee valuation the cost benefit for a fish from the first smolt
release (1000 rod caught level) would be 0.5. This means that for every pound invested a maximum of
50 pence would be returned. For succeeding releases the cost benefit would be 1.59. This benefit is
based on a maximum return of 4.5% and exploitation of 15%. It should be expected that the return
rate of 4.5% is highly unlikely and a return of zero probable in some years.
3.4. Timeline Based on the design and operational details produced to support the economic assessment, the
current timeline can be proposed. The Dee has historically been considered a MSW fishery. If stocking
were to go ahead, how long would it take until the first hatchery reared MSW fish return? First, it
should not be assumed that any fish will return; stocked fish have much lower survival rates than wild
fish (Aprahamian et al., 2003). Secondly, there is no guarantee that these fish will return as MSW
salmon (if at all). With those two caveats put to one side the following timeline is proposed:
1. Stocking actioned by the Dee board in November 2016 with one year to raise the funds;
2. Design and planning work would begin in November 2017 (18 months) and CAR and
broodstock licenses;
3. May 2019 - At least six months for planning application objections;
4. November 2020 – build begins (10 months);
5. September 2021 build finishes and broodstock collection begins;
6. Spring 2022 – first hatch;
7. Spring 2024 – stocking license application;
8. Spring 2025 – first hatchery smolt release;
9. 2026 – first returning grilse; and
10. 2027 – first returning MSW salmon.
3.5. Conclusion The stocking of smolts to improve the rod catch by 100 fish is not recommended on economic
grounds. This is because it will deliver little definitive benefit for significant cost and this number of
fish is a very small fraction of the rod catch (see Figure 3.1. below).
River Forth Fisheries Trust: River Dee Hatchery Assessment
22
Figure 3.1 - 100 fish as a % contribution to total catch
Secondly, even in the biologically unlikely event that 4.5% of smolts return, increasing the rod catch to
5000 is also difficult to justify. Leaving aside the huge numbers of broodstock required and assuming a
Dee Board operating budget similar to 2016 of £624,660 (DDSFB, 2016), an additional £200,000 would
be needed to operate the hatchery, along with the cessation of all other operations run by the Dee
Board and Trust. Funding the hatchery build would need almost £9 million. These figures also do not
take into account the loss of fish from natural production if the required broodstock were removed.
The 1000 level could be carried out if the Dee Board liquidated all its assets to fund 40% of the
hatchery build costs, with additional contributions making up the £2 million shortfall. Operating costs
for the hatchery reach 89% of the total 2016 Board operating costs.
Ultimately however, based on the broad range of survival rates in the literature there is no guarantee
that any fish will return and the investment will not be of any economic benefit to the Dee fishery or
any private individuals who would wish to donate to a hatchery.
The cost benefit assessment is based on the economics of building and running a hatchery only and
additional costs are possible. Rivers stocked with smolts might have lower long term catches than
rivers not stocked with smolts (Young, 2014). The potential loss to the naturally produced salmon rod
catch on the Dee from the short and long-term negative impacts of stocking and loss of broodstock
would need to be calculated and added to the cost benefit analysis to improve accuracy. It is possible
that other items would need to be added to the ‘cost’ side of the calculation (such as potential
impacts on the pearl mussel population and the reputational loss to the Dee of being a wild fishery)
and a specific in-depth economic review assessment (similar to that carried out on the Tweed) may be
required.
In addition to the other potential costs proposed above, stocking the Dee cannot simply be viewed as
an investment into the build and operation of a hatchery. The figures presented here show that
stocking would require the operating budget for bailiffing, management, angling development and
maintenance cut. (Staff salaries of over £250,000 for 2015 could not be maintained under any of the
three stocking scenarios).
This operating budget of the Board and Trust is also used to indirectly access much larger funds which
can support improvements on the Dee. A key task currently is the development of programmes which
bring in funding from central government, the EU and other agencies. (The Pearls in Peril project was
0
10
20
30
40
50
1952 1962 1972 1982 1992 2002 2012
% c
on
trib
itio
n
River Forth Fisheries Trust: River Dee Hatchery Assessment
23
originally judged to be worth £3.5 million, for an input of £100,000 from the Trust. In 2015 it
contributed £126,000 to the Dee Boards’ budget, a figure which represents over 30% of the entire levy
contribution.) Stocking potentially not only risks the viability of the salmon population on the Dee, but
also all the other activities carried out by the Board and Trust to ensure the continuation of the
fishery.
River Forth Fisheries Trust: River Dee Hatchery Assessment
24
4. Legal As mentioned throughout this report, there are a number of legal factors that could potentially
influence the decision to stock the Dee. They are:
The Dee SAC and other designations;
The Cairngorm National Park;
Licensing of issues relating to stocking;
The change to Fishery Management Organisations as part of the Wild Fisheries Review; and
The current legal obligation of the Board to protect, conserve and enhance the salmon fishery.
4.1. Protected sites Designated sites protect Scotland’s natural heritage through statute. There are different types of
designations, each conferring different levels of protection. The River Dee catchment contains
numerous protected sites ranging from regionally to internationally important areas. Site designations
of the River Dee catchment include:
Natura 2000 sites – European level designations established under the EC Habitats and Birds
Directives and including Special Areas of Conservation (SACs) and Special Protection Areas
(SPAs). SPAs protect species of birds while SACs protect other habitats and species of
community interest. Under the legislation, sites must be maintained in favourable condition.
The protection afforded by these designations is generally applied through planning and
licensing processes. If as a result of a planning application, activity or development there is
‘likely to be a significant effect’ upon the habitats or species featured in an SAC or SPA, then
an Appropriate Assessment (AA) must be obtained along with advice from Scottish Natural
Heritage as to whether the application can be accepted.
The establishment of a hatchery, because it may impact a notified feature of a site (Atlantic
salmon) would likely require several Appropriate Assessments (AA) to be carried out each time
a stocking event is carried out (C. Bean, SNH, pers. comm.). Both stages involved in the
process – the removal of adult broodstock and the stocking of hatchery reared fish into the
SAC – would have to be assessed against the conservation objectives of the SAC.
Sites of Special Scientific Interest – UK level designations under the Wildlife and Countryside
Act, 1981 and as updated. Provisions for protection and management of SSSis’ were provisions
recently improved by the Nature Conservation (Scotland) Act 2004 and the Wildlife and
Natural Environment (Scotland) Act 2010. Planning authorities must consider whether a
development will affect a site but they have considerable discretion in their decision making.
National Parks – Scotland level designations under the National Parks (Scotland) Act 2000 to
conserve countryside landscapes promote public enjoyment of them and promote sustainable
economic and social development.
River Forth Fisheries Trust: River Dee Hatchery Assessment
25
Ramsar sites – International designations under the Convention on Wetlands of International
Importance, 1971. All Ramsar sites in Scotland are also SPAs or SACs and their protection
comes primarily from those designations.
National Nature Reserves – In Scotland, these are declared by Scottish Natural Heritage. They
contain areas of important natural and semi natural terrestrial and coastal ecosystems and are
managed to conserve these habitats and to provide public recreation.
Local Nature Reserves – These can be declared by Local Authorities (following consultation
with the relevant statutory nature conservation agency). They are managed for conservation,
education and recreation.
Protected sites within the River Dee catchment There are many protected sites within the River Dee catchment. Many of these are SSSIs (Sites of
Special Scientific Interest), SACs (Special Conservation Areas) and Ramsar sites with a focus on upland
terrestrial habitats, wetlands, otters and standing waters. The whole of the upper catchment falls
within the Cairngorms National Park and there is substantial overlap with the Cairngorm SAC and the
Cairngorm Massif SPA. These protected sites give a picture characterising the catchment as a relatively
unspoiled, highland catchment with low nutrient waters and healthy fish populations – ideal
conditions for supporting high conservation value species such as Atlantic salmon and freshwater
pearl mussels. The primary designation of interest and relevance to the river and to these species is
the River Dee SAC.
The River Dee Special Area of Conservation There are 17 river SACs in Scotland that feature Atlantic salmon. The River Dee is one of 11 that have
Atlantic salmon as a primary qualifying reason for site selection.
The SAC includes the entire length of the main stem and almost all of the tributaries. Three interest
features form the basis for the designation, all of them species that are primary reasons for
designation. Habitats are not specifically mentioned in the designation (other than those supporting
the named species). The three species are:
Atlantic salmon (Salmo salar);
Freshwater pearl mussel (Margaritifera margaritifera); and
European otter (Lutra lutra).
Management of a salmon hatchery must consider the potential effects on all three interdependent
species. Otters are the top predator in this species complex, with salmon and freshwater pearl mussel
(FWPM) both used as food sources. However, with a wide range of alternative food sources available
to them, they may not be expected to respond substantially to the magnitude of changes in prey
composition that might result from any potential hatchery activity. Thus an AA may not necessarily be
required for otter. However, disturbance to a preferred prey species, for example, the European eel
undergoing a population crash, could make otters much more heavily dependent on salmonids for
survival, and could change this situation.
Sea trout, brown trout and Atlantic salmon are the only known hosts for the larval stage of pearl
mussel (Young and Williams, 1984). As 0+ salmon are generally more abundant than 0+ trout in most
River Forth Fisheries Trust: River Dee Hatchery Assessment
26
rivers, salmon are considered likely to be the most important hosts of the FWPM. However, there are
Scottish streams with few salmon yet they still support a mussel population, and these must be
considered to be largely trout dependent (Skinner et al., 2003). Salmon are the dominant species on
the River Dee, although there are also significant numbers of brown trout and a sea trout fishery. It is
thought that salmon are likely to be the preferred host of FWPM in the River Dee system, as is the
case for the neighbouring River Spey (I. Syme, SNH, pers. comm.), however, a study to confirm the
relative importance of the two species as FWPM hosts would be useful to better inform any
management decisions affecting either host species.
The long term survival of the FWPM depends ultimately on host availability and any significant
changes in salmon stocks could threaten FWPM populations (Cosgrove et al., 2000). Whilst juvenile
salmon densities throughout the River Dee system are generally well above thresholds for the
continued existence of FWPM (RAFTS, 2014 and Skinner et al., 2003), it is still necessary to guard
against localised impacts in areas where the salmon population may be augmented with stocking.
The relationship between salmon and FWPM is mutually beneficial. Adult mussels can filter up to 50
litres of water per day (Zuiganov et al., 1994 – in Skinner et al., 2003) which helps maintain water
quality at a level favourable to salmonids. In addition, FWPM beds may provide a microhabitat for the
invertebrates on which the salmon feed.
Conservation objectives The River Dee SAC brings with it conservation objectives stipulating what protection must be afforded
to the species identified in the designation, and to the structure and function of their supporting
habitats. The River Dee SAC aims ‘to avoid deterioration of the habitats of the qualifying species or
significant disturbance to the qualifying species, thus ensuring that the integrity of the site is
maintained and the site makes an appropriate contribution to achieving favourable conservation
status for each of the qualifying features’ (species).
In addition, the objectives are to ensure for the qualifying species that the following are maintained in
the long term:
1. Population of the species, including range of genetic types for salmon, as a viable component
of the site;
2. Distribution of the species within the site;
3. Distribution and extent of habitats supporting the species;
4. Structure, function and supporting processes of habitats supporting the species;
5. No significant disturbance of the species;
6. Distribution and viability of freshwater pearl mussel host species; and
7. Structure, function and supporting processes of habitats supporting freshwater pearl mussel
host species.
Implications of the conservation objectives for hatchery stocking
Objective 1 The main constraint to salmon management emerging from these conservation objectives is
maintaining genetic types for salmon as a viable component of the site (objective 1). This requirement
has profound implications for management, particularly with regard to the use of a hatchery. Atlantic
River Forth Fisheries Trust: River Dee Hatchery Assessment
27
salmon have been shown to exhibit genetic structuring between river systems (Webb et al., 2007; King
et al., 2007), tributaries and even within tributaries. This genetic differentiation indicates that
salmonids become adapted to their local conditions and benefit by homing to their natal areas where
their offspring are likely to have optimal survival rates (Fraser et al. 2011). Inherited characteristics
may include growth rate, run-timing (Stewart et al., 2006), variation in age at smolting (Englund et al.,
1999), sea-age at maturity (Niemela, 2006), disease resistance and survival behaviours. The
introduction of life stages from a hatchery which have not been subject to the natural selection
pressures of the stocked area would likely lead to reduced survival within the local population
(McGinnity et al., 2003). See Section 1 for further details.
The River Dee was included in a recent project looking at genetic structuring within a range of Scottish
rivers (the FASMOP project8). It was found that fish could be correctly assigned to one particular
tributary of the River Dee – the Water of Dye – 90% of the time (Coulson et al., 2013). This indicates a
high level of unique adaptation to that particular tributary. Stocking the Water of Dye with hatchery
reared fish using broodstock from elsewhere in the River Dee may have a profound negative effect on
survival rates. The management implications of these findings are therefore that hatcheries must be
tributary (or preferably reach) specific. There is a strong likelihood that this structuring will be found
elsewhere on the catchment. For this reason, only broodstock from a particular tributary or section of
main stem should be used to stock that area. This will be very difficult to achieve, as although an adult
may be taken from a certain section of the river for artificial breeding, unless it is taken from a redd
site, there is no guarantee that is the watercourse where that fish will breed (and even then, this fish
may move to spawn in another location (Taggart et al., 2001). Practically, removing sufficient
broodstock for a tributary based stocking operation is a significant investment in time and capital and
will require a large team operating across the catchment.
Restocking a river with hatchery reared stock from elsewhere has been successful in some cases, and
where there is no existing population (as per guidance from RAFTS, 2014). (However, evidence from
the Tyne and Thames is that stocking simply maintained the fishery while straying from elsewhere
rebuilt the natural population – Milner et al., 2004; Griffiths et al., 2011). The Dee is currently in
favourable condition for both adult stocks and juvenile densities (RAFTS, 2014) using the SAC
monitoring procedures. Any hatchery on the River Dee system must therefore use only tributary or
reach-specific local broodstock and be strictly controlled to avoid damaging the salmon populations
that are already in place.
In addition to location of broodstock, the number of breeders must also be considered as having a
profound effect on the genetic heterogeneity and health the population. A large number of adult
broodstock are required to avoid inbreeding. 50 mating pairs, chosen at random, should be used as
broodstock (Bailey & Kincaid, 1989). The River Tay hatchery maintains a broodstock of around 300
fish, based on reconditioning kelts.
Other principles, drawn from Mills (1991), that should be adhered to when managing a hatchery to
preserve genetic integrity include the following:
8 Focussing Atlantic Salmon Management on Populations. See RAFTS website at
http://www.rafts.org.uk/focussing-atlantic-salmon-management-on-populations/
River Forth Fisheries Trust: River Dee Hatchery Assessment
28
Broodstock should not be kept in captivity for more than one generation and should be
obtained from the wild each year that stocking is to take place;
Collection of spawners should encompass the entire spawning migration period;
Some pristine stocks should be maintained and conserved in each geographical area;
One-on-one matings should be the norm wherever possible, i.e. males are only used once; and
Robust monitoring must be carried out to ensure the stocking programme is having the
desired outcome.
A post stocking monitoring plan is imperative to determine whether the hatchery is contributing to an
increase in returning adult salmon available for the rod fishery, whether it is delivering value for
money and most importantly - not causing harm to existing stocks. Broodstock must be genetically
typed so that returning hatchery produced fish can be identified in the relevant fisheries. This will
determine the effect of stocking on rod catches. Annual electrofishing of stocked areas is necessary to
determine the level of survival of hatchery juveniles (in the event freshwater stages are stocked). In
areas where naturally spawned juveniles are also present, genetic testing will determine the relative
abundance of stocked and natural juveniles and determine the extent to which one replaces the
other. This is of particular importance on the River Dee as it is likely that stocking will take place in
areas where natural populations are also present, and damage to the natural population must be
avoided at all costs and to comply with the conservation objectives of the SAC designation.
Objective 2 Objective 2 states that the current distribution of the species within the river system should be
maintained. Stocking should not therefore be carried out in a way that raises the likelihood of causing
detriment or decline in the natural salmon population.
Detailed analysis of fry densities in seven tributaries of the River Dee and one main stem site during
the most recent round of Site Condition Monitoring revealed a ‘high variability in fry density across the
catchment, possibly reflecting local differences in juvenile productivity’. This could point to some
tributaries having degraded or unsuitable habitat or insufficient numbers of spawners (However,
yearly variation may not indicate long term patterns of decline or increase (see Sections 1 and 2).
Stocking in areas that already have a head of salmon present is not effective and can do more harm
than good (see Section 2.1).
It has been established that sufficient numbers of local broodstock should be used to stock a given
area. This may give rise to a logistical problem of obtaining sufficient broodstock from an area of lower
productivity. There is also increased risk associated with removing rare broodstock from an area
where the population is already under pressure. This should only be attempted if there is strong
evidence to suggest that productivity and survival would be higher with the hatchery in place. Adults
found in sub optimal areas may not necessarily breed there, instead moving off to more suitable areas
to breed, where spawning success and juvenile survival would be higher. In addition, males that may
have spawned multiple times would be unable to do so if removed to a hatchery (where they are to be
used only once). Both of these scenarios would cause stocking to result in a lowering of natural
productivity.
To address the problem of removing adults from a degraded area, mended kelts can be reconditioned
following spawning and used as broodstock (as in the Tay). There is significant cost associated with
River Forth Fisheries Trust: River Dee Hatchery Assessment
29
reconditioning kelts in an appropriate installation, particularly as they can only be used once in order
to avoid inbreeding, therefore new kelts must be reconditioned each year.
Objective 3 Distribution and extent of habitats supporting the species. Hatchery and stocking activities are unlikely
to have any impact on habitats.
Objective 4 Structure, function and supporting processes of habitats supporting the species. Hatchery and
stocking activities are unlikely to have any impact on the structure, function and supporting process of
habitats.
Objective 5 No significant disturbance of the species. As mentioned above, hatchery activities are unlikely to
disturb otters as they have a wide range of preferred and alternative food sources they can utilise. The
intention of stocking would be to increase the number of salmon in the River Dee system, but as
discussed above, unless hatchery activities are very stringently controlled and strict principles are
adhered to, there is the potential to cause damage to the existing populations, to reduce genetic
fitness and survival. Any such negative impact would no doubt have a follow on effect upon the FWPM
that are dependent on their salmon hosts.
Objective 6 Distribution and viability of freshwater pearl mussel host species. Since the host species, juvenile
salmon, are also protected as a feature of the SAC designation, the impact of a potential hatchery on
their distribution has already been addressed in objective 2 above.
Objective 7 Structure, function and supporting processes of habitats supporting freshwater pearl mussel host
species. Since the host species, juvenile salmon, are also protected as a feature of the SAC designation,
the impact of a potential hatchery on the structure and function of their habitat has already been
addressed in objective 4 above.
Other Atlantic salmon SACs in Scotland There are 17 SACs in Scotland that include Atlantic salmon as part of their designation. Some are also,
like the River Dee, designated for freshwater pearl mussel and otter. Each species can be considered
either a primary or a secondary reason for designation, and this will affect the level of protection
accorded to the species and its supporting habitats and has implications for management. The species
designation for each of the 17 SACs is shown in Table 4.1.
River Forth Fisheries Trust: River Dee Hatchery Assessment
30
Table 4.1 - Atlantic salmon SACs in Scotland, showing salmon, freshwater pearl mussel and otter as
primary or secondary reason for designation (marked 1 or 2 accordingly, or – where the species is
not included)
Special Area of Conservation Atlantic salmon
Freshwater pearl mussel
Otter
Berriedale and Langwell Waters 1 - -
River Bladnoch 1 - -
River Borgie 2 1 2
River Dee 1 1 1
Endrick Water 2 - -
Langavat 1 - -
Little Gruinard River 1 - -
River Moriston 2 1 -
River Naver 1 1 -
North Harris 2 1 -
River Oykel 2 1 -
River South Esk 1 1 -
River Spey 1 1 1
River Tay 1 - 2
River Teith 2 - -
River Thurso 1 - -
River Tweed 1 - 1
Of the 17 designated sites, only the River Spey SAC has the same range and level of interest features
as the River Dee SAC. The two sites have the highest possible tier of protection compared to the other
SACs, with the same three interdependent species as primary reasons for designation.
The Spey Fishery Board runs a hatchery operation that now observes many of the principles discussed
above e.g. local source of broodstock and genetic testing of broodstock for comparison with the rod
fishery. The hatchery is targeted primarily at areas with low salmon productivity due to restricted
access rather than degraded habitat, for example, upstream of impassable or partially passable man
made barriers (areas upstream of natural barriers are not considered for stocking as this would disturb
the unique gene pools of long-established and isolated brown trout populations). Fish are released as
0+ fry and electrofishing monitoring of juveniles and genetic sampling of broodstock and returning
adults are carried out to determine the effectiveness of the hatchery and its contribution to the
number of returning adults. Stocking activity on the Spey has become more stringent, reduced and
targeted over time, and it is acknowledged by the Spey Fishery Board in their 2015 annual report that:
‘A more sustainable strategy, that will benefit the whole river, is to conserve stocks to ensure there are
adequate fish available to spawn, and to ensure that the habitat in the nursery areas is as good as
possible, so as to promote enhanced survival through the parr and ultimately smolt stages of the
salmon life cycle.’
Genetic analysis of rod caught fish in 2010 and 2011, compared to hatchery released juveniles of four
and five years previously, showed that hatchery fish contributed less than 2% to the rod fishery (SFB
River Forth Fisheries Trust: River Dee Hatchery Assessment
31
Annual Report, 2013). The Board note that since broodstock were taken from the wild, they are likely
to have spawned naturally in any case, and so this 2% is not a direct measure of hatchery contribution.
The actual additional contribution from the hatchery activity is likely to be considerably less than this
(if any). These figures indicate that a hatchery, run appropriately, constitutes a significant monetary
outlay for what is likely to be a vanishingly small gain.
During discussion, Colin Bean of SNH queried the case for stocking the Dee based on recent poor
catches. Reiterating the conclusions from the 2013 IBIS workshop on stocking that it should not be a
tool of first use, the evidence base for stocking was raised. Secondly, it was stated that stocking could
be a potential long term risk and that the activity may not be compliant with conservation objectives
of the SAC.
Therefore, two appropriate assessments would be required (for broodstock removal and for stocking)
and it should be shown that the number of potential eggs removed during broodstock gathering
should be at least matched by the stocked return. However, matching is unlikely to prove a sound
basis of the economic argument for stocking.
Cairngorm National Park was also interviewed with respect to stocking. Prior efforts at stocking the
Dee would have occurred before CNG was established. Therefore there was no planning panel when
it came to construction of a hatchery. Currently, if it is deemed that any planning permissions were
needed for the construction of the hatchery, access routes to the hatchery or associated structures,
the planning decision could be sent to the Park board. The representative of the Park stated that the
Dee was crucial to the Park but that a balance should be struck between the economic and
conservation aims of the park. Any stocking process should review all the factors but other avenues,
such as improved catchment and land management would be a start point for the CNG before
stocking. It was stated that there is value in a feasibility study but there would be a certain amount of
concern over stocking and that physical habitat is paramount.
SNH are of the view the stocking is a potential risk to the Conservation Objectives of the SAC, and
legislation requires that the scale and impact of such risks must be fully assessed before such activities
can be carried out. To allow such an assessment to be carried out, supporting information on adult
and juvenile numbers will be required, as well as sound rationale for carrying out this activity.
Another licensee outwith the Dee board is Marine Scotland Science. MSS will need to grant a license
for broodstock removal and Brian Shaw of the Spey has stated that MSS are applying greater scrutiny
to that Board’s stocking license applications every year, and that a stocking panel is convened every
year within the Board to continually reassess the need for stocking internally. Therefore it is likely that
there will be a significant input from the Dee Board and Trust to stocking applications and, if stocking
were to be given the go ahead, this does not guarantee that permission will be granted in future years.
4.2. Legal Obligations of the board: Management of salmon The management of Atlantic salmon is currently within the remit of the Dee District Salmon Fishery
Board. However, following the Wild Fisheries Review (Scottish Government 2014) and the publication
of a consultation paper on draft provisions for a Wild Fisheries (Scotland) Bill and Strategy (February
2016), it is abundantly clear that the current structure of management will cease. A new hierarchical
structure will instead be activated, comprising a national Fisheries Management Unit (FMU) dealing
with strategy and other high level operations, with a number of local Fisheries Management
River Forth Fisheries Trust: River Dee Hatchery Assessment
32
Organisations (FMOs) delivering practical management of all wild fish species in the designated FMO
areas. This will mean that powers currently within the sphere of the Board may be given to the
National Management Unit. This may include the license to stock Atlantic salmon. Therefore, prior to
instigating any formal license application studies, the final position of responsibility for stocking within
the new management system should be clarified.
An in-detail description of the legal requirement of the Board, the relevant legislation and upcoming
changes is presented in Appendix E
4.3. Legal obligations of the Board: Alternatives to stocking The factors which influence the survival and productivity of Atlantic salmon are numerous.
Subsequently, any actions which can mitigate negative pressures can boost salmon productivity.
These negative pressures can come in two forms; natural ecosystem pressures such as predation and
anthropogenic pressures. The purpose of this report is not to investigate limiting ecosystem pressures
and limiting populations of species such as seal, dolphin, otters, cormorant, and other piscivorous
birds should not be considered acceptable fishery management practice. This does raise an issue as
the majority of stakeholders responded that predation by dolphins, seals, and birds were a pressure.
(Carss et al., 1990, suggested that predation, in this case by otters, had little impact on the Dee.) It
will be difficult for the Dee Board to persuade a regulator that predator control is a viable
management strategy.
Netting As part of the changes to fishery management in Scotland, coastal netting is to temporarily cease. The
Dee board annual report 2015 stated that although the netting operations on the Dee have practically
stopped, 18% of net catch from a neighbouring catchment were Dee fish. Therefore, by closing these
netting stations the Dee will see an increase in returning salmon this year. 2015 catch returns from
fixed engine netting on the Esk were 2803 salmon (MSS catch statistics). By closing this fishery an
additional 504 salmon would have returned to the Dee. Based on the exploitation rate of 15% on the
Dee, this could have corresponded to an additional 75 rod caught fish. With all coastal netting station
closed as part of the new Conservation Regulations, it is most probable that a related increase of
returning fish could supply the additional 100 rod caught requested by stocking. Therefore, the lower
level increase in rod catch proposed by the Dee Trust will be met by changes in salmon management.
This cessation of coastal netting will last for three years, and seeking an extension of this closure
should be a Dee Board (or FMO) priority.
Fish passage, habitats and morphology. Figures 4.1 to 4.3 (fish passage, morphology and water quality respectively) show the range of
pressures on the Dee system as determined by SEPAs WFD monitoring (SEPA, 2015). Areas considered
to have good or better quality are highlighted in blue and green, other colours represent reduced
quality. These figures have been reproduced from the SEPA River Basin Management Plan (RBMP).
With the exception of water quality, it is clear that work remains to bring the catchment up to an
acceptable standard for habitats and fish migration.
River Forth Fisheries Trust: River Dee Hatchery Assessment
33
Figure 4.1 - SEPA quality scoring for fish migration on the Dee (blue – high quality, green – good
quality, yellow – moderate, orange – poor, red – bad)
Figure 4.2 - SEPA quality scoring for physical condition (blue – high quality, green – good quality,
yellow – moderate, orange – poor, red – bad)
Figure 4.3 – SEPA quality scoring for water quality (blue – high quality, green – good quality, yellow
– moderate, orange – poor, red – bad)
It is commonly believed that the issue of weirs has been managed on the Dee catchment (stakeholder
survey respondents). However, there are 149 individual barriers which remain, 11 of which are still
impassable (see Figure 4.4. from the Dee walkover). While it is often tempting to think of weirs
exerting a pressure only on upstream passage of fish, they also prevent the formation of new habitats,
especially gravel habitats. Another factor, mentioned previously, is that fish passes do not operate as
effectively as believed. Very little peer-reviewed information is available on this; however Noonan et
al., 2012 found that maximum passability ranged from only 70% to 20% (dependent on the type with
less natural-type passes proving less passable). There was also a major impact on juveniles, with 30%
of attempted downstream passes unsuccessful. If this pattern of effectiveness is applicable to the Dee,
then clearly there will still be a major impact on the migratory patterns of salmon.
River Forth Fisheries Trust: River Dee Hatchery Assessment
34
Figure 4.4 - Manmade obstacles on the Dee
The cumulative impact of weirs, whether deemed passable with fish passage or without, should bear
further investigation. Weirs of only 30cm have been shown to retard migration (Gerlier and Roche,
1998). Even with facilitation of passage, factors such as increased catchability, opportunistic
predation, disease, parasitism and increased susceptibility to pollution can cause mortality (de Leaniz,
2008) and reduced breeding success (see Figure 4.5 for an incidence of disease outbreak at a weir with
an operational fish pass). Often, the simple presence of a weir (even with passage) can cause a
cessation of migration (Solomon et al., 1999). Instream barriers also prevent new gravels from being
transported downstream, with old redds having reduced survival of ova when compared with new
ones (Wheaton et al., 2004). Therefore it is too simplistic and probably incorrect to state that the
issue of barriers has been solved on the Dee and, in general, there are still significant knowledge gaps
when it comes to understanding the impacts of weirs of fish migration (Milner et al., 2013).
River Forth Fisheries Trust: River Dee Hatchery Assessment
35
A B
Figure 4.5 - Two of approximately 100 salmon killed during a low flow event at Westfield weir on
the Allan water during 2016. A functioning fish pass is present on the structure.
River Restoration In addition to the impact from weirs, the abundance, diversity and relative presence of instream
habitat plays a role in limiting salmonid populations. Habitat complexity and wood in rivers have been
shown to increase juvenile survival (Quinn & Peterson., 1996); elevate fry density (Venter et al., 2008,
Floyd et al., 2008 ); and increase spawning (Floyd et al., 2008). (Conversely an absence of wood is
linked to reduce growth rates of juvenile salmon (Finstad et al., 2007)).
Recent work by SEPA on the Dee highlighted that there still remain significant areas of damage to
instream and bank structure and habitats on the Dee, including the mainstem. These are shown
below in Figure 4.6. Ten of these channels have been selected for SEPA funded restoration work to
improve instream habitat.
Figure 4.6 – Areas of reduced physical habitat quality (morphology) on the Dee (from cbec, 2011)
River Forth Fisheries Trust: River Dee Hatchery Assessment
36
In addition to these areas, many areas of the upper Dee and its tributaries have historically been
affected by tree removal or grazing pressure preventing tree growth (see Figure 4.7).
Figure 4.7 – The upper Dee near the Linn of Dee
This can lead to a reduction in the number of pools (Montgomery et al., 1998), biological refuges
(Gurnell et al., 2005) and increase sediment input into channel (Haigh et al., 2004) making the channel
less suitable for older juvenile fish such as parr. Although the Dee Trust are currently planting
significant number of trees in the catchment, it will take many years before those trees mature and
become established and the benefit of these trees can be mimicked by placing wood directly into
rivers.
One of the results of Storm Frank was to reveal considerable lengths of previously unrealised river
modifications on the channel (H. Moir, pers. comms). These works, which are possibly related to the
Muckle spate (1829), will have resulted in a lack of lateral channel movement over a significant period
of time. Although often seen as a negative from the human perspective, channel movement is
important in creating new habitats and recycling gravels trapped in the floodplain and events such as
the storms experienced in 2015/16 are often critical in the long term evolution and function of rivers.
An indication of the static and straightened condition of the Dee at Bielside is illustrated in Figure 4.8.
River Forth Fisheries Trust: River Dee Hatchery Assessment
37
Figure 4.8 – Evolution of the Dee channel
Calculations based on maps are often crude, as actual river wetted area will change on any given day,
and older maps often less accurate. However, the reduced diversity of the channel from the mid
1800s is clearly demonstrated, perhaps reflecting channel modifications related to extreme flood
events (H. Moir, pers. comm). Additionally, based on these maps the total wetted area from 1750 to
the 2000s may have decreased by a half, while the decrease from the 1850s to the 2000s is by about a
third. Quite simply, less river area equals less production and on the Dee catchment there are clearly
significant lengths of channel that continue to be constrained by human influences.
Habitat restoration is now a common tool for returning to a more functioning natural state, and,
experience with stream restoration for the enhancement of salmonid populations in northwest USA
suggest that, when projects are implemented correctly, sizeable increases in fish production can be
achieved (Roni et al., 2010). However, making direct links between channel restoration or
enhancement activity and improved fish stocks and fishery performance is often difficult (Roni et al.,
2008). They reported that a poor understanding of fish stocks before restoration, limited scope and
poor monitoring made it difficult to fully assess the benefits. They did provide evidence that barrier
removal or mitigation measures and instream habitat enhancement actions increased local fish
populations.
River Forth Fisheries Trust: River Dee Hatchery Assessment
38
The effectiveness of restoration at improving fish populations was reviewed by Kemp, (2010).
Improvements varied enormously and were influenced by many other factors (Armstrong et al., 2003);
life-stage and seasonal changes in habitat use (Heggenes & Gunnar Dock, 2001), the size of the rivers
restored (Miller et al., 2008), and the relatively small scale of the interventions9 (e.g. Roni et al., 2002).
As a result, many programmes report inconclusive results on fish populations or may appear to have
failed to demonstrate any improvement. In a review presented at the River Restoration Centre
conference in 2013 by McDermott, out of 320 hundred reviewed restoration projects, less than five
conclusively proved an ecological benefit. This was not due to the lack of benefits but more simply
because of the poor quality of, or even complete absence of, a monitoring programme.
Despite over-ambitious expectations, limited scales of application and inadequate monitoring
programmes (Liermann and Roni, 2008) some examples do exist where positive responses in salmonid
production have resulted from habitat intervention. Nislow et al., (1999) have shown that the
introduction of instream structures to increase the abundance of preferred habitats led to increased
retention of juvenile salmon during the critical first summer period. Gargan et al., (2002) reported
improvements to the populations of Atlantic salmon parr in reaches of 13 Irish streams following
creation of in-stream structures to restore substrate and flow complexity. Jong et al., (1999) show a
positive response in Atlantic salmon parr abundance in response to instream structures placed into
Newfoundland streams, attributing increases to improved habitat complexity reducing competition.
Solazzi et al. (2000) provide evidence of a doubling of smolt production in Coho salmon (Oncorhynchus
kisutch) in the four years following establishment of in-stream structures to improve habitat quality in
small catchments ( <20km2) in the Pacific northwest.
In Scotland, two highly visible programmes on the Spey and Tweed have demonstrated the benefit of
river restoration. Sections of the Allt Lorgy (Spey) and Eddleston (Tweed) were restored with salmonid
spawning observed in the very first winter following completion and fish densities potentially
responding to this restoration (salmon fry densities are higher and on a distinct upward trajectory
when compared with a control location – however this pattern requires further investigation and
validation with further years’ data – Spey Foundation, unpublished) restoration efforts in the Nith
District have been complicated by stocking. However, it is possible that habitat improvement has
been the paramount influence on population recovery (D. Parke, pers. comm). While these examples
may demonstrate increased incidence of spawning or fish densities, their benefit to the rod catch
remains to be quantified.
The cost of restoration can be high. However in areas where willing landowners, space, and an
absence of major infrastructure is present (like areas of the upper catchment) it can be much cheaper
than stocking, would be much more likely to gain regulatory approval and would be significantly more
beneficial to fish. The Allt Lorgy project, which restored a significant stretch of channel above
Carrbridge, cost in the region of £50,000 with sea trout observed spawning in the new channel two
9It is worth remembering that in many cases these works are aiming to undo decades or even centuries of
damage to river systems – it is unsurprising that improvements cannot be detected by a small number of electrofishing samples.
River Forth Fisheries Trust: River Dee Hatchery Assessment
39
weeks after completion. The Eddleston, which has the complicating factor of a rail line and road cost
somewhat more at approximately £650,00010 (Luke Comins, pers. comm).
River restoration has been, and continues to be, the key management tool on the Dee for improving
smolt output and should be seen as the primary mechanism for boosting juvenile salmon production
and survival. Although certain high visibility projects such as the Culter Dam mitigation, catchment
riparian planting and Pearls in Peril demonstrate the breath of projects; significantly more works are
carried out. Currently there are over 150km of buffer strips being actively managed and maintained,
varying between six and over 100 metres in width. Major croy and embankment removal projects
have been delivered in the last four years, while over 30 barriers have been eased in the last 10 years.
Fencing (specifically developed to exclude livestock, deer, or both) to protect developing riparian
habitats is being increased in partnership with landowners yearly, and almost 1000 tree enclosures
have been created. These works are ongoing and deliver an increasing protective cushion against
highly intensive land management practises. As these schemes develop, they will also deliver habitat
benefits to the river. However, as discussed these will take time to mature, and it is possible that the
benefits of the tree planting won’t be maximised for 30 years. The restoration works carried out on
the Dee are internationally recognised, with the CASS (Conservation of Atlantic Salmon) project, which
included significant works on the Dee, winning the ‘Best of Best’ LIFE Nature award in 2010.
10
Corresponding to 2km of remeanders, 66ha of new riparian woodland, 16km of fences, 19 wetlands and 89 flow restrictors using wood. That project also was responsible for preventing Peebles from flooding during Storm Desmond. Heavy rainfall in the winter of 2015/16 raised the river to a level which had previously flooded the town. In that one event the Eddleston restoration is likely to have returned a significant portion of its capital investment.
River Forth Fisheries Trust: River Dee Hatchery Assessment
40
5. Discussion The stocking of Atlantic salmon has been a cornerstone of fisheries management and salmon
population management across the Atlantic for decades. The extent of stocking across the salmon’s
home range is remarkable, and the complete inventory of rivers stocked (and where the fish used to
stock these fish came from) is an unknown for many older programmes. Currently there is a
consensus that salmon are in decline (MSS, 2016; Bacon et al., 2015) and the Dee echoes this. But the
decline is not a consistent pattern across all stock components (MSS, 2016) and stocking may not yet
be the appropriate management response. The Dee SAC is in favourable condition status and the
available information (e.g. SNH, 2011) suggests that the Dee population is not at risk (C. Bean, P.
McGinnity and S. McKelvey, pers. comm), therefore any current efforts to stock the Dee should be
considered fishery enhancement stocking.
The now mature body of information that stocking of salmon is counterproductive to both the fishery
(Young, 2013) and salmon populations (RAFTS, 2014, Young, 2013 – and numerous references therein)
should have provided a strong case for the cessation of fishery enhancement stocking in Scotland.
This case has been made so forcefully that practitioners of stocking are now reviewing their entire
datasets in light of this information (for example, a current project on the River Bush) and Natural
Resources Wales banned all stocking of hatchery reared Atlantic salmon in 2015. There is also a
developing acceptance that what has worked on one river may not work on another (The Tay District
Salmon Fishery Board has stated that releasing smolts on that system is not warranted currently and
unlikely to reap benefits for the fishery when compared with the Ranga (TDSFB, 2011)).
There has been a failure in the translation of this scientific advice into a form that stakeholders can
accept. It is true that the concepts underpinning the science are often only understood by the
practitioners of that individual field; however this is not an excuse. More effort should be made to
ensure the central message is understood and accepted. However, it should also be accepted that the
scientific work forming the basis for stocking advice is often complex and the information within
primary sources may not always be accessible to individuals (including other scientists and managers)
lacking the formal training in genetic and statistical methods.
Part of the problem is caused by a failure to translate the message into terms that are comparable
with the contrasting argument. Stocking is tangible; young fish can be observed growing in hatcheries
and being placed into the river system. They can be seen swimming off and there is a definitive and
immediate effect observed. When this is coupled with the highly variable and cyclical nature of adult
returns, it is no surprise that many (including many well respected fishery managers and scientists)
find it so attractive to link a year of high return with a stocking programme. When we counter this
with scientific arguments based on genetic fitness, insufficient monitoring programmes (‘’more study
is needed’’), the relevance of statistical probability and the reluctance of scientists to make firm
unequivocal conclusions based on what in other walks of life would be deemed certainty, it is clear
why the argument for stocking remains. But this failure by scientists to communicate the results of
their work effectively does not mean it is incorrect.
Further complicating the matter is the fact that, in some isolated but very well publicised occasions,
fishery enhancement stocking appears to have worked. The Icelandic Ranga is the prime example.
However, it should be abundantly clear that comparing the Dee and the Ranga is not a practical
exercise. The Ranga is an exceptional case, atypical of even Icelandic rivers, and due to its geographic
River Forth Fisheries Trust: River Dee Hatchery Assessment
41
location, its physical state and the land through which it flows it is certainly not comparable to an east
coast Scottish river like the Dee. The unsuitability of the river for salmon spawning coupled with a
remarkable difference in both return rates and catchability of Ranga fish puts it apart from all rivers.
Without stocking there would be no salmon in that system and it is being completely and artificially
sustained by ranching. This has come at the cost of concern regarding the straying of these fish
(Isakson et al., 1997).
Stocking also occurs in rivers such as the Tay and Spey. While these rivers are located close to the
Dee, they are structurally different in terms of catchment size, tributary sizes and geology, and
reviewing preliminary rod catches on those rivers for 2016 (www.fishpal.com) clearly highlights the
differences in the fisheries. (The Spey is enjoying a bumper year; catches on the Tay are somewhat
suppressed despite the high Pitlochry dam fish count; the Dee catch is an improvement on 2015).
And, most importantly, it appears that stocking does not result in any significant positive impact on
rod catch on those rivers; in fact, the scientific evidence suggests that stocking may be harming the
salmon stock of those rivers. In rivers in Scotland where stocking has been deemed successful they
occur either in rivers so impacted by human pressure that the stocking programme could be deemed
conservation stocking with a major fishery benefit (the Conon), or that the actual and unequivocal
effect of stocking operations has yet to be formally verified (the Carron – Curran et al., 2015).
A rivers’ size is another consideration when reviewing potentially successful stocking programmes.
Stocking seems to be most successful in small, accessible Atlantic rivers where angling exploitation
rates can be higher (as there is much less river to cover and fish are concentrated into a small number
of small pools). Translating any perceived stocking success from these river types to the Dee, with its
vast areas of channel, is simply not feasible (S. McKelvey, P. McGinnity, E. Verspoor, pers. comm). And
again, based on figures presented in Stewart et al., (2015) the question remains whether stocking on
any of the big east coast Scottish rivers has contributed anything other than a small token return to
the rod.
There is absolutely no doubt that Atlantic salmon are undergoing a critical period as a species in
Scotland. Marine mortality rates are increasing (MSS, 2010) with rising sea temperatures affecting
post-smolt survival (Friedland et al., 2005); growth of adults at sea (Friedland & Todd; 2012); and the
number and size of returning adults (NASCO, 2012). But stocked salmon are much less likely to survive
the marine environment than naturally reproduced salmon (MSS, 2015; Aprahamian et al.; 2011) and
survival by stocked fish during periods of low sea survival is relatively less than during high survival
phases (McGinnity et. al., 2004). Naturally produced salmon are more likely to survive this current
period of oceanic uncertainty and therefore it is this wild-born stock which should be supported, most
effectively through ensuring habitat quantity, quality and diversity can support them across the range
of climactic conditions and anthropogenic pressures.
Therefore fishery enhancement stocking in a system which is still healthy (although potentially
declining), extraordinarily large, and with an overarching impact from outwith the river environment is
not appropriate. It is not appropriate because firstly there is a mature body of information that it is
harmful to a still functioning population, that those natural fish who can be most impacted by stocking
are the best equipped to deal with increasing problems at sea and that the other methods of boosting
fish populations, especially river restoration and returning the river to a more natural state have not
River Forth Fisheries Trust: River Dee Hatchery Assessment
42
been exhausted. The fact that stocking operations may have worked elsewhere simply does not
translate to the Dee.
Set against the challenges that now face salmonids, which are often linked to global climate change,
the only real effort that can be made by fishery managers is within their individual river systems.
Other than stocking, quick win options are limited. The role of barriers in rivers has been discussed,
but this discussion should be formally expanded to include not only upstream fish migration, but the
role of these structures in retaining gravels for spawning and altering natural flow patterns. Because
many of the barriers on the Dee have been made passable it is assumed that their impact has been
negated. This might not be accurate. It is also clear that the number of barriers recorded during the
Dee walkover has come as a surprise, and this data should be formally reviewed in depth.
It is a similar case with river restoration. The channel evolution map in Section 4 has provided an
example of the changes in the river over time and should hint at the scale of task needed to restore
the river back to anything approaching a more natural state. One of the reasons for the lack of
measureable responses in fish communities is that river restoration has not been appropriately scaled
and although significant efforts have already been made on the Dee in this area more needs to be
done. For example, a common response among stakeholders when asked about additional tree
planting is that it has been ’done’. Nonetheless a quick drive through the upper Dee and tributaries
will demonstrate the lack of tree cover; there are still enormous lengths of watercourse to be
reforested. Where trees have been planted, they are still young and immature and this point leads to
a critical component of river restoration – it takes time; sometimes a lot of time. While this message
must be communicated better, river restoration and physical habitat improvement should be
considered the most appropriate approach to increase long term and stable improvements in smolt
output.
Broadly speaking the stakeholder responses were evenly split between support for stocking and not
stocking. Among those who supported it there was some indication that that they would support an
increase in costs to support stocking, but this was not unanimous. Other key issues which were raised
were that predation was a major driver of declines and catches had declined to such an extent that
fishing effort was no longer related to catch. The responses from the stakeholders demonstrated the
expected well informed and engaged nature of the group. This survey was very limited in its
application and its purpose was simply to gauge the breath of opinion. A more inclusive survey will
probably be necessary to formalise these opinions.
While stakeholder survey responses show that uptake of days on the Dee is commonly at between 25
to 50%, there is a belief that reduced angling effort is not linked to reduced catches. This may be true
but seems counter-intuitive and it (the assumed link between catch and population) needs tested.
Claiming that reduced catches definitely demonstrate a collapse in salmon stocks while also claiming
those catches are independent of fishing effort is a contrary view, and may be seen as such by
regulators. Even if reduced effort only contributes to a minor reduction in overall catch, this reduction
needs some quantification.
It should still be considered that new and exciting research is continuously emerging about salmon
productivity in healthy catchments. A good example of this is research looking at the loss of nutrients
from catchments and that recovering and retaining these nutrients could increase and perhaps even
double fish production (Williams et al., 2010; Williams et al., 2009; Nislow et al., in press). Fishery
River Forth Fisheries Trust: River Dee Hatchery Assessment
43
enhancement stocking cannot even begin to approach such increased productivity on a river like the
Dee and during a phone discussion it was stated that the authors feel the maximum productivity boost
has not yet been realised (S. McKelvey, pers. comm). However; once again the necessities of scientific
rigour have made it difficult to establish these responses with absolute certainty at this current time.
This report looked at three strands of evidence to decide on stocking. The first was the biological. The
evidence from the published sources has been presented (Section 2), and the problems with
translating that message discussed. The second part of the biological argument was the review of
data from electrofishing and habitat walkovers. These data have been collected for a number of
reasons, and the differing purposes are evident. Ultimately the dataset was assessed as insufficient to
provide the required level of information to inform on the need for stocking.
As a first measure, a baseline data collection programme of fixed locations with fixed methods is
required, and the temptation to move to different locations (or methods) in response to short term
pressures avoided. The current catchment health monitoring programme being carried out by the
Dee Trust should be considered the answer, however it will take time to develop this dataset to the
point where it can be used to inform on any future conservation needs. With regard to enumerating
adult returns, although the walkover data does not identify a main stem weir low down, potential
locations should be investigated for a fish counter to operate.
The second strand, the economic appraisal, has demonstrated that the cost to stock the Dee is
enormous. Although there may be a weak argument for stocking on economic grounds of at the 1000
level, three factors should be considered when accepting this at face value:
The estimated start-up costs exceed £3.5 million and the yearly running costs exceed
£500,000. Even considering an enormous donation, this is a huge outlay and there is no
guarantee of licenses to stock in future years;
The survival/return values represent a range. While these are taken from the literature the
lowest figure for return rates is not uncommonly zero. Operation of a hatchery cannot
guarantee a return; and
The cost benefit assessment does not include the loss of natural productivity caused by the
removal of broodstock and their progeny.
The final strand of investigation related to the legal obligations of the board to the fishery and the SAC
designation of the Dee. These are very well covered in the relevant section and in the Appendices.
Considering the level of protection given to the Dee, the granting of a license to stock salmon under
the current scenario is highly unlikely and would be an extremely resource intensive data collection
exercise. However, even if these data already existed and regulators were amenable to fishery
enhancement stocking, the changes to the management structures for wild fishing in Scotland mean it
would be prudent to wait until this process has been resolved and the body ultimately responsible for
the application identified before proceeding. There are still a wide range of other options open to the
Board to assist the salmon population on the river. The proactive approach to restoration has initiated
a trajectory of habitat improvement on the river, but the scale of habitat improvement needs to be
expanded enormously and these efforts will need time.
River Forth Fisheries Trust: River Dee Hatchery Assessment
44
5.1. What next? Even if broodstock collection was actioned this year and a mothballed hatchery of suitable capacity
existed, it might be the early 2020s before the first runs return. Given that an immediate start is
unlikely to meet licensing, planning and animal welfare requirements, multi sea-winter returns would
probably not be seen until the latter half of the 2020s.
Although this report has identified numerous biological, economic and legal constraints to fishery
enhancement stocking, there are scenarios under which stocking may gain regulatory approval. These
are catastrophic crashes in the Dee population or sub-catchment sub-populations, most likely to be
related to marine survival, of a scale that conservation stocking is required. There should be a plan
for this event outlined within a new Dee stocking policy and should include a procedure for dealing
with any private license applications, to the Board or future FMO, for stocking of salmon.
The first and most important piece of this plan should be the continued development of a dataset (the
catchment health monitoring) to support the detailed catch records and data from the Girnock and
Baddoch burns. This dataset can be used to determine the health of the population and should be
linked to the SAC site conditions assessments (i.e. reviewed at regular intervals). This will provide cost
benefit to both SNH and the future Dee FMO. The datasets should be regularly reviewed from a
statistical perspective to ensure any conclusions taken are justified, and survey methods should be
fixed. A first specific avenue of investigation could be the relationship between fry and 1+ parr.
A complete inventory of the genetic differentiation within the Dee catchment should be established
with individual subpopulations identified, and plans to remove broodstock from those areas
developed and suitable storage facility should be identified. Some of this information on the genetic
composition of the Dee stock(s) already exists (Coulson et al., 2013), however spatial coverage is low
(16 sites) and it was unclear if the observed lack of diversity related to the methods used or true
genetic similarity.
Discussions should be held with all the relevant licensing authorities to establish a way to expedite
hatchery operations in the event of a crash. Ultimately, although this report has found that the time
for stocking has not yet arrived, the Dee should have access to a hatchery to use for conservation
stocking. Therefore, an inventory of mothballed hatcheries should be made along with a formal
determination of production capacities and identification of the upgrades required to bring these
operations on-line.
Efforts to increase the level of smolt output should continue and should concentrate on increasing the
parr carrying capacity in the river. This should involve a catchment review of the more subtle impacts
of in-channel barriers, along with an independent review of the restoration approaches on the
catchment. The catchment heath monitoring dataset should be used to understand the catchment
scale effect of these efforts and quantify the benefit to the rod.
5.2. Conclusions The consensus of this report, based on the reviewed data and other information developed within is
that it is currently not appropriate to stock the Dee with Atlantic salmon of any life history stage.
This is because:
River Forth Fisheries Trust: River Dee Hatchery Assessment
45
The link between catch and population size is not conclusive and there are still knowledge
gaps relating to the population size and juvenile production and survival;
The only available evidence suggests smolt productivity is stable;
Catches have declined, but there has also been a major decline in efforts to catch salmon;
Stocking of smolts is excessively expensive;
Stocking is seen as a potential risk to the conservation status of the Dee SAC and possibly
contrary to the objectives of the Cairngorm National Park and a risk to the natural population;
A license activity is not guaranteed and cannot be supported with currently available data;
There are still numerous other options available to potentially improve smolt output including
further habitat improvements (with appropriate time being for those improvements to
mature) and efforts to reduce the impact of weirs;
There will be an increase in returning fish is 2016 due to the closure of coastal netting in
Scotland;
The upcoming changes in salmon management mean it is not clear who would regulate,
authorise, fund and be responsible for stocking;
Other information based on analysis of the biological data and on more detailed data sets from the
Girnock and Baddoch burns show that there may not be an issue with natural production of smolts,
and that numbers of returning adults have not reached the level at which this smolt production level is
at risk. Further work should be carried out to determine the applicability of this theory across the Dee
catchment.
River Forth Fisheries Trust: River Dee Hatchery Assessment
46
6. References Anderson JH, Faulds PL, Atlas WI, & Quinn TP. (2013). Reproductive success of captively bred and
naturally spawned Chinook salmon colonizing newly accessible habitat. Evolutionary Applications 6:
165–179.
Aprahamian MW, Smith K.M, McGinnity P, McKelvey S &Taylor J. (2003). Restocking of salmonids –
opportunities and limitations. Fisheries Research. 62:211-227
Araki H, Cooper B & Bloui, MS. (2007). Genetic effects of captive breeding cause a rapid, cumulative
fitness decline in the wild. Science 318:100–103.
Araki H, Berejikian B A, Ford MJ, Blouin MS. (2008). Fitness of hatchery-reared salmonids in the wild.
Evolutionary Applications 1:342–355.
Araki H, Cooper B, Blouin & M S. (2009). Carry-over effect of captive breeding reduces reproductive
fitness of wild-born descendants in the wild. Biology Letters 5:621–624.
Armstrong JD, Kemp PS, Kennedy GJA, Ladle M and Milner, NJ. (2003). Habitat requirements of
Atlantic salmon and brown trout in rivers and streams. Fisheries research. 62:143-170
Bailey J & Kincaid HL. (1989). Fish Breeding Principles In: Atlantic Salmon Broodstock Management
and Breeding Handbook. Ed Kincaid H L & Stanley JG. pp8
Berejikian BA, Tezak EP, Flagg TA, LaRae AL, Kummerow E, Mahnken CVW. (2000). Social dominance,
growth, and habitat use of age-0 steelhead (Oncorhynchus mykiss) grown in enriched and
conventional hatchery rearing environments. Canadian Journal of Fisheries and Aquatic Sciences
57:628–636.
Bilsby H. (2006). The River Dee: Stock Component Review 2006. Report prepared for The River Dee
Trust.
Bacon PJ, Malcolm IA, Fryer RJ, Glover RS, Millar CP & Youngson AF. (2015). Can conservation stocking
enhance juvenile emigrant production in wild Atlantic salmon? Transactions of the American Fisheries
Society
Bourret V, O’Reilly PT, Carr JW, Berg PR & Bernatchez L. (2011). Temporal change in genetic integrity
suggests loss of local adaptation in a wild Atlantic salmon (Salmo salar) population following
introgression by farmed escapees. Heredity 106: 500–510.
Carss DN, Kruuk H & Conroy JWH. (1990). Predation on adult Atlantic salmon, Salmo salar L., by otters,
Lutra lutra L., within the River Dee system, Aberdeenshire, Scotland. Journal of Fish Biology. 37:935-
944
Coulson MW, Laughton R, Shaw B, Armstrong A & Verspoor E. (2013). The use of genetic parentage
analysis to assess hatchery contribution of Atlantic salmon on the River Spey. Report prepared for the
Focusing Atlantic Salmon Management on Population
Coulson M, Hawkins L, Hudson A, Armstrong A, Cauwelier E, Stradmeyer L, Gilbey J, Sinclair C &
Verspoor E. (2013) An Overview of Population Genetic Structuring in the River Dee. Marine Scotland
Science Report 03/13.
River Forth Fisheries Trust: River Dee Hatchery Assessment
47
Crozier WW & Kennedy GJA. (1994). Application of semi-quantitative electrofishing to juvenile
salmonid surveys. Journal of Fish Biology. 45:159-164
Curran MJ, Kindness R, Verspoor E., Coulson M, Tipping R, Minting P, Tosney J, Duncan A & Smith M.
(2015). Project Report: River Carron Restoration Project. Rivers and Lochs Institute, University of
Highlands and Islands.
Englund V, Niemela E, Lansman M & Heino M. (1999). Variations in Atlantic salmon, Salmo salar L.,
smolt age in tributaries of the River Teno, Finland. Fisheries Management and Ecology. 6: 83-86.
Fleming IA & Einum S. (1997). Experimental tests of genetic divergence of farmed from wild Atlantic
salmon due to domestication. ICES Journal of Marine Science. 54:1051-1063
Floyd TA, MacInnes C & Taylor BR. (2008). Effects of artificial woody structures on Atlantic salmon
habitat and populations in a Nova Scotia stream. River Research and Applications. 25:272-282
Fraser DJ. (2008). How well can captive breeding programs conserve biodiversity? A review of
salmonids. Evolutionary Applications. 1:535-586
Fraser DJ, Weir LK, Bernatchez L, Hansen MM & Taylor EB. (2011). Extent and scale of local adaptation
in salmonid fishes: review and meta-analysis. Heredity. 106:404-420
Finstad AG, Einum S, Forseth T & Ugedal T. (2007). Shelter availability affects behaviour, size-
dependent and mean growth of juvenile Atlantic salmon. Freshwater Biology. 52:1710-1718
Friedland, KD, Reddin, DG, McMenemey, JR & Drinkwater, KF. (2003). Multidecadal trends in North
American Atlantic salmon (Salmo salar) stocks and climate trends relevant to juvenile survival.
Canadian Journal of Fisheries and Aquatic Science. 60. 563-583.
Friedland KD, Chaput G & MacLean JC. (2005). The emerging role of climate in post-smolt growth of
Atlantic salmon. ICES Journal of Marine Science. 62:1338-1349
Freidland KD & Todd CD. (2012). Changes in Northwest Atlantic Arctic and Subarctic conditions and
the growth response of Atlantic salmon. Polar Biology. 35:593-609
Gargan P, O’Grady M, Delanty K, Igoe F & Byrne C, (2002). The effectiveness of habitat enhancement
on salmon and trout stocks in streams in the Corrib Catchment. In Proceedings of the 13th
International Salmonid Habitat Enhancement Workshop. Edited by M. O’Grady. Central Fisheries
Board, Dublin, Ireland (pp. 220-223)
Gerlier M & Roche P. (1998). A radio A radio telemetry study of the migration of Atlantic salmon
(Salmo salar L.) and sea trout (Salmo trutta trutta L.) in the upper Rhine. Hydrobiologia. 371:283
Godfrey JD. (2005). Site condition monitoring of Atlantic salmon SACs. Report by the SFCC to Scottish
Natural Heritage
Griffiths AM, Jonathan S, Ellis JS, Clifton-Dey D, Machado-Schiaffino G, Bright D, Garcia-Vazquez E &
Stevens JR. (2011). Restoration versus recolonisation: The origin of Atlantic salmon (Salmo salar L.)
currently in the River Thames. Biological Conservation. 144: 2733-2738
River Forth Fisheries Trust: River Dee Hatchery Assessment
48
Gurnell A, Tockner K & Edwards P. (2005). Effects of deposited wood on biocomplexity of river
corridors. Frontiers in Ecology and the Environment. 3:377-382.
Gurney WSC, Bacon PJ, Tyldesley G, & Youngson AF. (2008). Process-based modelling of decadal
trends in growth, survival, and smolting of wild salmon (Salmo salar) parr in a Scottish upland stream.
Canadian Journal of Fisheries and Aquatic Science. 65:2606-2622.
Hastie LC & Young MR. (2003). Timing of spawning and glochidial release in Scottish freshwater pearl
mussel (Margaritifera margaritifera) populations. Freshwater Biology. 48: 2107-2117
Hawkins LA, Magurran AE & Armstrong JD. (2008). Ontogenetic learning of predator recognition in
hatchery-reared Atlantic salmon, Salmo salar. Animal Behaviour. 75:1663–1671.
Heggenes J & Gunnar Dokk J, (2001). Contrasting temperatures, waterflows, and light: seasonal
habitat selection by young Atlantic salmon and brown trout in a boreonemoral river. Regulated Rivers:
Research & Management. 17:623-635.
Hendry K, Sambrook H & Waterfall R. (2007). Assessment of salmon stocks and the use of
management targets; a case study of the River Tamar, England. Fisheries management and Ecology.
14:7-19.
Hutchings JA & Fraser DJ. (2008). The nature of fisheries- and farming-induced evolution. Molecular
Ecology 17:294–313.
Isakson A, Oskarsson SM, Einarsson SM, & Jonasson J. (1997). Atlantic salmon ranching: past
problems and future management. ICES Journal of Marine Science. 54:1188-1199
Johannsson M, Isaksson A, EIIidason Th, & Osakarsson S. (1996). Maintenance of angling through
smolt releases in the Ranga river in southern Iceland. ICES CM. 1996/M:6
Jong MVZD, Cowx IG & Scruton DA. (1997). An evaluation of instream habitat restoration techniques
on salmonid populations in a Newfoundland stream. Regulated Rivers-Research and Management.
13:603-614
Jonsson N, Jonsson B & Hansen LP. (2003) Marine survival and growth of wild and released hatchery-
reared Atlantic salmon. Journal of Applied Ecology. 40: 900-911.
Jonsson B & Jonsson N. (2006). Cultured Atlantic salmon in nature: a review of their ecology and
interaction with wild fish. ICES Journal of Marine Science: Journal du Conseil. 63:1162–1181.
Kallio-Nyberg I, & Koljonen ML. (1997). The genetic consequence of hatchery-rearing on life-history
traits of the Atlantic salmon (Salmo salar L): a comparative analysis of sea-ranched salmon with wild
and reared parents. Aquaculture 153:207–224.
Kemp P. 2010. In-channel placement of structures to enhance habitat complexity and connectivity for
stream-dwelling salmonids. In Salmonid Fishes: Freshwater habitat management. ed. P. Kemp .
Blackwell
Liermann M & Roni P. (2008). More sites or more years? Optimal study design for monitoring fish
response to watershed restoration. North American Journal of Fisheries Management. 28:935-943.
River Forth Fisheries Trust: River Dee Hatchery Assessment
49
Montgomery DR & Buffington JM. (1998). Channel process, classification, and response potential. In:
R.J. Naimen and R.E. Bilby (Ed.) River Ecology and Management, pp. 13-42. Springer-Verlag, New York.
Miller DJ, Burnett KE & Benda LE. (2008). Factors controlling availability of spawning habitat for
salmonids at the basin scale. American Fisheries Society Symposium. 65:103-120.
McLean JE, Seamons TR, Dauer MB, Bentzen P & Quinn TP. (2008). Variation in reproductive success
and effective number of breeders in a hatchery population of steelhead trout (Oncorhynchus mykiss):
examination by microsatellite-based parentage analysis. Conservation Genetics. 9:295–304.
McClure MM, Utter FM, Baldwin C, Carmichael RW, Hassemer PF, Howell PJ, Spruell P, Cooney TD,
Schaller HA. & Petrosky CE. (2008). Evolutionary effects of alternative artificial propagation programs:
implications for viability of endangered anadromous salmonids. Evolutionary Applications 1: 356–375.
McGinnity P, Prodöhl P, Maoileidigh NO, Hynes R, Cotter D, Baker N, O’Hea B & Ferguson A. (2004).
Differential lifetime success and performance of native and non-native Atlantic salmon examined
under communal natural conditions. Journal of Fish Biology. 65 (Supplement A): 173-187.
McGinnity P, Prodohl P, Ferguson A, Hynes R, Maoileidigh NO, Baker N, Cotter D, O’Hea B, Cooke D,
Rogan G, Taggart J, & Cross TF. (2003). Fitness reduction and potential extinction of wild populations
of Atlantic salmon, Salmo salar, as a result of interactions with escaped farm salmon. Proceedings of
the Royal Society of London Series B – Biological Sciences. 270:2443-2450.
Metcalfe NB, Valdimarsson SK, Morgan IJ. (2003). The relative roles of domestication, rearing
environment, prior residence and body size in deciding territorial contests between hatchery and wild
juvenile salmon. Journal of Applied Ecology 40:535–544.
Mills CPR & Piggins DJ. (1983). The release of reared salmon smolts (Salmo salar) into the Burrishoole
River System (Western Ireland) and their contribution to the rod and line fishery. Aquaculture
Research. 14:165-175
Milner NJ, Russell IC, Aprahamiam M, Inverarity R, Shelley J & Rippon P. (2004). The role of stocking in
the recovery of the River Tyne salmon fisheries. Environment Agency Fisheries Technical Report
No2004/1.
Milner NJ, Harrison AH, and McDermott T. (2013). Review of studies into salmonid migration patterns
and responses to river flows. Report prepared by APEM for SNIFFER
Milot E, Perrier C, Papillon L, Dodson JJ, & Bernatchez L. (2013). Reduced fitness of Atlantic salmon
released in the wild after one generation of captive breeding. Evolutionary Applications. 6: 472–485.
Niemela E, Erkinaro J, Julkunen M, Hassinen E, Lansman M & Brors S. (2006) Temporal variation in
abundance, return rate and life histories of previously spawned Atlantic salmon in a large subarctic
river. Journal of Fish Biology. 68:1222-1240.
Nislow KH, Folt CL & Parrish DL. (1999) Favorable foraging locations for young Atlantic salmon:
application to habitat and population restoration. Ecological Applications. 9:1085-1099.
Noonan MJ, Grant JWA & Jackson CD. (2011). A quantitative assessment of fish passage efficiency.
Fish and Fisheries. 13:450-464
River Forth Fisheries Trust: River Dee Hatchery Assessment
50
Postle M & Moore L. (1996). Economic evaluation of recreational fisheries. In: P Hickley & H Tompkins
(eds) Recreational Fisheries; Social, Economic and Management Aspects. Oxford: Fishing News Books,
Blackwell Science, pp. 184–199.
Poulet N. (2007). Impact of weirs on fish communities in a Piedmont stream. River Research and
Applications. 23:1034-1047
Radford A, Riddington G, Anderson J & Gibson H. 2004. The economic impact of game and coarse
angling in Scotland. Research report prepared for the Scottish Executive, Edinburgh. 66pp
Rivers and Fisheries Trusts of Scotland. (2014). Data Supporting Site Condition Monitoring of Atlantic
salmon SACs. Scottish Natural Heritage Commissioned Report No. 755.
Quinn GP & Keough MJ. (2002). Experimental design and data analysis for biologists. Cambridge
University Press
Roni P, Beechie TJ, Bilby RE, Leonetti FE, Pollock MM & Pess GR. (2002). A review of stream restoration
techniques and a hierarchical strategy for prioritizing restoration in Pacific Northwest watersheds.
North American Journal of Fisheries Management. 22:1-20.
Roni P, Hanson K & Beechie T. (2008). International review of the effectiveness of stream
rehabilitation. North American Journal of Fisheries Management. 28:856-890.
Roni P, Pess G, & Morley S. (2010). Monitoring salmon stream restoration: guidelines based on
experience in the American Pacific northwest. In Salmonid Fishes: freshwater habitat management. ed.
P. Kemp . Blackwell
RSPCA. 2015. RSPCA welfare standards for farmed Atlantic salmon.
Spey Fishery Board. (2013). Annual Report 2013
Spey Foundation. (2015). Spey mainstem post spate and stocking fry counts Sept 2014
Scottish Environment Protection Agency. (2015). River basin management plan for the Scotland River
Basin District:2015-2027. Scottish Government.
Scottish Natural Heritage (2011). Data Supporting Site Condition Monitoring of Atlantic salmon SACs.
Scottish Natural Heritage Commissioned Report No. 755.
Simpson I. (2003). Inventory of adult fish counters and traps in Scotland 2003. Fisheries Research
Services Internal Report No 18/03.
Skinner A, Young M, & Hastie, L. (2003) Ecology of the Freshwater Pearl Mussel. Conserving Natura
2000 River Ecology Series No. 2, English Nature, Peterborough.
Solazzi MF, Nickelson TE, Johnson SL & Rodgers JD. (2000). Effects of increasing winter rearing habitat
on abundance of salmonids in two coastal Oregon streams. Canadian Journal of Fisheries and Aquatic
Sciences. 57:906-914
Stewart DC, McKelvey S & Armstrong JD. (2015). Ranching to the rod: an evaluation of adult returns
from hatchery-reared Atlantic salmon smolts released in Scottish rivers. Scottish Marine and
Freshwater Science Report. 6.
River Forth Fisheries Trust: River Dee Hatchery Assessment
51
Stewart DC, Middlemas SJ & Youngson AF. 2006. Population structuring in Atlantic salmon (Salmo
salar): evidence of genetic influence on the timing of smolt migration in sub-catchment stocks. Ecology
of Freshwater Fish. 15: 552-558.
Solomon DJ (1985). Salmon stock and recruitment, and stock enhancement. Journal of Fish Biology.
27:45-57
Solomon DJ, Sambrook HT & Broad KJ. (1999). Salmon Migration and River Flow: Results of Tracking
Radio tagged Salmon in Six Rivers in South West England. Environment Agency R&D Publication,
Bristol 4:110.130.
SQW. (2015). Economic impact from angling on the Tweed river system.
Spey Fishery Board. (2013). Annual Report 2013
Taggart JB, McLaren LS, Hay DW, Webb JH, & Youngson, A.F. (2001). Spawning success in Atlantic
salmon (Salmo salar L.): a long-term DNA profiling based study conducted in a natural stream.
Molecular Ecology. 10:1047-1060.
Tay District Salmon Fishery Board. (2011). Could smolt stocking achieve similar results for the Tay as in
Iceland? Frequently asked questions No.2.
Thériault, V., Moyer, G. R., Banks, M. A. (2010). Survival and life history characteristics among wild and
hatchery coho salmon (Oncorhynchus kisutch) returns: how do unfed fry differ from smolt releases?
Canadian Journal of Fisheries and Aquatic Sciences 67:486–497.
Thériault V, Moyer GR, Jackson LS, Blouin MS & Banks MA. (2011). Reduced reproductive success of
hatchery coho salmon in the wild: insights into most likely mechanisms. Molecular Ecology 20:1860–
1869.
The River Dee Trust. (2013). River Dee stock component review 2013.
The River Dee Trust. (2016). 2015 Dee Stock Assessment.
Thorpe JE & Mitchel KA. (1981) Stocks of Atlantic salmon (Salmo salar) in Britain and Ireland:
discreteness and current management. Canadian Journal of Fisheries and Aquatic Science 38: 1576-90.
Uusitalo L, Kuikka S, Romakkaniemi A. (2005). Estimation of Atlantic salmon smolt carrying capacity of
rivers using expert knowledge. Ices Journal of Marine Science. 62:708-722.
Venter O, James WA, Noel MV & Kim J. (2007). Mechanisms underlying the increase in young-of-year
Atlantic salmon (Salmo salar) density with habitat complexity. Canadian Journal of Fisheries and
Aquatic Sciences. 65:1956-1964
Williamson KS, Murdoc AR, Pearsons TN & Ford MJ. (2010). Factors influencing the relative fitness of
hatchery and wild spring Chinook salmon in the Wenatchee River, Washington. Canadian Journal of
Fisheries and Aquatic Sciences 67:1840–1851.
Young KA. (2013). The balancing act of captive breeding programmes: salmon stocking and angler
catch statistics. Fisheries management and Ecology. 20:434-444
River Forth Fisheries Trust: River Dee Hatchery Assessment
52
Young MR & Williams J. (1984) The status and conservation of the freshwater pearl mussel in Great
Britain. Biological Conservation. 25:35-52.
Zuiganov V, Zotin A, Nezlin L & Tretiakov V. (1994) The freshwater pearl mussels and their relationships
with salmonid fish. VNIRO, Moscow. 104 pp.
River Forth Fisheries Trust: River Dee Hatchery Assessment
53
Acknowledgements The Project team wishes to thank all the correspondents and stakeholders who took the time to
respond to requests for information, data and opinion during the course of the work. All input,
whether quoted or not has proved invaluable to the formation of this report.
Top Related